diff --git a/quantstats/__init__.py b/quantstats/__init__.py
index 1b17699a..c6c4b32f 100644
--- a/quantstats/__init__.py
+++ b/quantstats/__init__.py
@@ -25,7 +25,7 @@
from . import stats, utils, plots, reports
-__all__ = ['stats', 'plots', 'reports', 'utils', 'extend_pandas']
+__all__ = ["stats", "plots", "reports", "utils", "extend_pandas"]
# try automatic matplotlib inline
utils._in_notebook(matplotlib_inline=True)
@@ -102,7 +102,9 @@ def extend_pandas():
_po.treynor_ratio = stats.treynor_ratio
_po.probabilistic_sharpe_ratio = stats.probabilistic_sharpe_ratio
_po.probabilistic_sortino_ratio = stats.probabilistic_sortino_ratio
- _po.probabilistic_adjusted_sortino_ratio = stats.probabilistic_adjusted_sortino_ratio
+ _po.probabilistic_adjusted_sortino_ratio = (
+ stats.probabilistic_adjusted_sortino_ratio
+ )
# methods from utils
_po.to_returns = utils.to_returns
@@ -146,4 +148,6 @@ def extend_pandas():
_po.plot_monthly_heatmap = plots.monthly_heatmap
_po.metrics = reports.metrics
+
+
# extend_pandas()
diff --git a/quantstats/_plotting/core.py b/quantstats/_plotting/core.py
index ef256a77..4a29f499 100644
--- a/quantstats/_plotting/core.py
+++ b/quantstats/_plotting/core.py
@@ -19,6 +19,7 @@
# limitations under the License.
import matplotlib.pyplot as _plt
+
try:
_plt.rcParams["font.family"] = "Arial"
except Exception:
@@ -27,57 +28,93 @@
import matplotlib.dates as _mdates
from matplotlib.ticker import (
FormatStrFormatter as _FormatStrFormatter,
- FuncFormatter as _FuncFormatter
+ FuncFormatter as _FuncFormatter,
)
import pandas as _pd
import numpy as _np
import seaborn as _sns
from .. import (
- stats as _stats, utils as _utils,
+ stats as _stats,
+ utils as _utils,
)
-_sns.set(font_scale=1.1, rc={
- 'figure.figsize': (10, 6),
- 'axes.facecolor': 'white',
- 'figure.facecolor': 'white',
- 'grid.color': '#dddddd',
- 'grid.linewidth': 0.5,
- "lines.linewidth": 1.5,
- 'text.color': '#333333',
- 'xtick.color': '#666666',
- 'ytick.color': '#666666'
-})
+_sns.set(
+ font_scale=1.1,
+ rc={
+ "figure.figsize": (10, 6),
+ "axes.facecolor": "white",
+ "figure.facecolor": "white",
+ "grid.color": "#dddddd",
+ "grid.linewidth": 0.5,
+ "lines.linewidth": 1.5,
+ "text.color": "#333333",
+ "xtick.color": "#666666",
+ "ytick.color": "#666666",
+ },
+)
-_FLATUI_COLORS = ['#FEDD78', '#348DC1', '#BA516B', '#4FA487', '#9B59B6',
- '#613F66', '#84B082', '#DC136C', '#559CAD', '#4A5899']
-_GRAYSCALE_COLORS = ['#000000', '#222222', '#555555', '#888888', '#AAAAAA',
- '#CCCCCC', '#EEEEEE', '#333333', '#666666', '#999999']
+_FLATUI_COLORS = [
+ "#FEDD78",
+ "#348DC1",
+ "#BA516B",
+ "#4FA487",
+ "#9B59B6",
+ "#613F66",
+ "#84B082",
+ "#DC136C",
+ "#559CAD",
+ "#4A5899",
+]
+_GRAYSCALE_COLORS = [
+ "#000000",
+ "#222222",
+ "#555555",
+ "#888888",
+ "#AAAAAA",
+ "#CCCCCC",
+ "#EEEEEE",
+ "#333333",
+ "#666666",
+ "#999999",
+]
def _get_colors(grayscale):
colors = _FLATUI_COLORS
- ls = '-'
- alpha = .8
+ ls = "-"
+ alpha = 0.8
if grayscale:
colors = _GRAYSCALE_COLORS
- ls = '-'
+ ls = "-"
alpha = 0.5
return colors, ls, alpha
-def plot_returns_bars(returns, benchmark=None,
- returns_label="Strategy",
- hline=None, hlw=None, hlcolor="red", hllabel="",
- resample="A", title="Returns", match_volatility=False,
- log_scale=False, figsize=(10, 6),
- grayscale=False, fontname='Arial', ylabel=True,
- subtitle=True, savefig=None, show=True):
+def plot_returns_bars(
+ returns,
+ benchmark=None,
+ returns_label="Strategy",
+ hline=None,
+ hlw=None,
+ hlcolor="red",
+ hllabel="",
+ resample="A",
+ title="Returns",
+ match_volatility=False,
+ log_scale=False,
+ figsize=(10, 6),
+ grayscale=False,
+ fontname="Arial",
+ ylabel=True,
+ subtitle=True,
+ savefig=None,
+ show=True,
+):
if match_volatility and benchmark is None:
- raise ValueError('match_volatility requires passing of '
- 'benchmark.')
+ raise ValueError("match_volatility requires passing of " "benchmark.")
if match_volatility and benchmark is not None:
bmark_vol = benchmark.loc[returns.index].std()
returns = (returns / returns.std()) * bmark_vol
@@ -87,7 +124,9 @@ def plot_returns_bars(returns, benchmark=None,
if isinstance(returns, _pd.Series):
df = _pd.DataFrame(index=returns.index, data={returns.name: returns})
elif isinstance(returns, _pd.DataFrame):
- df = _pd.DataFrame(index=returns.index, data={col: returns[col] for col in returns.columns})
+ df = _pd.DataFrame(
+ index=returns.index, data={col: returns[col] for col in returns.columns}
+ )
if isinstance(benchmark, _pd.Series):
df[benchmark.name] = benchmark[benchmark.index.isin(returns.index)]
if isinstance(returns, _pd.Series):
@@ -98,32 +137,37 @@ def plot_returns_bars(returns, benchmark=None,
df = df.dropna()
if resample is not None:
- df = df.resample(resample).apply(
- _stats.comp).resample(resample).last()
+ df = df.resample(resample).apply(_stats.comp).resample(resample).last()
# ---------------
fig, ax = _plt.subplots(figsize=figsize)
- ax.spines['top'].set_visible(False)
- ax.spines['right'].set_visible(False)
- ax.spines['bottom'].set_visible(False)
- ax.spines['left'].set_visible(False)
+ ax.spines["top"].set_visible(False)
+ ax.spines["right"].set_visible(False)
+ ax.spines["bottom"].set_visible(False)
+ ax.spines["left"].set_visible(False)
# use a more precise date string for the x axis locations in the toolbar
- fig.suptitle(title, y=.94, fontweight="bold", fontname=fontname,
- fontsize=14, color="black")
+ fig.suptitle(
+ title, y=0.94, fontweight="bold", fontname=fontname, fontsize=14, color="black"
+ )
if subtitle:
- ax.set_title("%s - %s \n" % (
- df.index.date[:1][0].strftime('%Y'),
- df.index.date[-1:][0].strftime('%Y')
- ), fontsize=12, color='gray')
+ ax.set_title(
+ "%s - %s \n"
+ % (
+ df.index.date[:1][0].strftime("%Y"),
+ df.index.date[-1:][0].strftime("%Y"),
+ ),
+ fontsize=12,
+ color="gray",
+ )
if benchmark is None:
colors = colors[1:]
- df.plot(kind='bar', ax=ax, color=colors)
+ df.plot(kind="bar", ax=ax, color=colors)
- fig.set_facecolor('white')
- ax.set_facecolor('white')
+ fig.set_facecolor("white")
+ ax.set_facecolor("white")
try:
ax.set_xticklabels(df.index.year)
@@ -135,9 +179,11 @@ def plot_returns_bars(returns, benchmark=None,
# ax.fmt_xdata = _mdates.DateFormatter('%Y-%m-%d')
# years = sorted(list(set(df.index.year)))
if len(years) > 10:
- mod = int(len(years)/10)
- _plt.xticks(_np.arange(len(years)), [
- str(year) if not i % mod else '' for i, year in enumerate(years)])
+ mod = int(len(years) / 10)
+ _plt.xticks(
+ _np.arange(len(years)),
+ [str(year) if not i % mod else "" for i, year in enumerate(years)],
+ )
# rotate and align the tick labels so they look better
fig.autofmt_xdate()
@@ -145,9 +191,8 @@ def plot_returns_bars(returns, benchmark=None,
if hline is not None:
if not isinstance(hline, _pd.Series):
if grayscale:
- hlcolor = 'gray'
- ax.axhline(hline, ls="--", lw=hlw, color=hlcolor,
- label=hllabel, zorder=2)
+ hlcolor = "gray"
+ ax.axhline(hline, ls="--", lw=hlw, color=hlcolor, label=hllabel, zorder=2)
ax.axhline(0, ls="--", lw=1, color="#000000", zorder=2)
@@ -156,11 +201,12 @@ def plot_returns_bars(returns, benchmark=None,
_plt.yscale("symlog" if log_scale else "linear")
- ax.set_xlabel('')
+ ax.set_xlabel("")
if ylabel:
- ax.set_ylabel("Returns", fontname=fontname,
- fontweight='bold', fontsize=12, color="black")
- ax.yaxis.set_label_coords(-.1, .5)
+ ax.set_ylabel(
+ "Returns", fontname=fontname, fontweight="bold", fontsize=12, color="black"
+ )
+ ax.yaxis.set_label_coords(-0.1, 0.5)
ax.yaxis.set_major_formatter(_FuncFormatter(format_pct_axis))
@@ -194,14 +240,31 @@ def plot_returns_bars(returns, benchmark=None,
return None
-def plot_timeseries(returns, benchmark=None,
- title="Returns", compound=False, cumulative=True,
- fill=False, returns_label="Strategy",
- hline=None, hlw=None, hlcolor="red", hllabel="",
- percent=True, match_volatility=False, log_scale=False,
- resample=None, lw=1.5, figsize=(10, 6), ylabel="",
- grayscale=False, fontname="Arial",
- subtitle=True, savefig=None, show=True):
+def plot_timeseries(
+ returns,
+ benchmark=None,
+ title="Returns",
+ compound=False,
+ cumulative=True,
+ fill=False,
+ returns_label="Strategy",
+ hline=None,
+ hlw=None,
+ hlcolor="red",
+ hllabel="",
+ percent=True,
+ match_volatility=False,
+ log_scale=False,
+ resample=None,
+ lw=1.5,
+ figsize=(10, 6),
+ ylabel="",
+ grayscale=False,
+ fontname="Arial",
+ subtitle=True,
+ savefig=None,
+ show=True,
+):
colors, ls, alpha = _get_colors(grayscale)
@@ -210,8 +273,7 @@ def plot_timeseries(returns, benchmark=None,
benchmark.fillna(0, inplace=True)
if match_volatility and benchmark is None:
- raise ValueError('match_volatility requires passing of '
- 'benchmark.')
+ raise ValueError("match_volatility requires passing of " "benchmark.")
if match_volatility and benchmark is not None:
bmark_vol = benchmark.std()
returns = (returns / returns.std()) * bmark_vol
@@ -232,45 +294,52 @@ def plot_timeseries(returns, benchmark=None,
returns = returns.last() if compound is True else returns.sum()
if isinstance(benchmark, _pd.Series):
benchmark = benchmark.resample(resample)
- benchmark = benchmark.last(
- ) if compound is True else benchmark.sum()
+ benchmark = benchmark.last() if compound is True else benchmark.sum()
# ---------------
fig, ax = _plt.subplots(figsize=figsize)
- ax.spines['top'].set_visible(False)
- ax.spines['right'].set_visible(False)
- ax.spines['bottom'].set_visible(False)
- ax.spines['left'].set_visible(False)
+ ax.spines["top"].set_visible(False)
+ ax.spines["right"].set_visible(False)
+ ax.spines["bottom"].set_visible(False)
+ ax.spines["left"].set_visible(False)
- fig.suptitle(title, y=.94, fontweight="bold", fontname=fontname,
- fontsize=14, color="black")
+ fig.suptitle(
+ title, y=0.94, fontweight="bold", fontname=fontname, fontsize=14, color="black"
+ )
if subtitle:
- ax.set_title("%s - %s \n" % (
- returns.index.date[:1][0].strftime('%e %b \'%y'),
- returns.index.date[-1:][0].strftime('%e %b \'%y')
- ), fontsize=12, color='gray')
-
- fig.set_facecolor('white')
- ax.set_facecolor('white')
+ ax.set_title(
+ "%s - %s \n"
+ % (
+ returns.index.date[:1][0].strftime("%e %b '%y"),
+ returns.index.date[-1:][0].strftime("%e %b '%y"),
+ ),
+ fontsize=12,
+ color="gray",
+ )
+
+ fig.set_facecolor("white")
+ ax.set_facecolor("white")
if isinstance(benchmark, _pd.Series):
ax.plot(benchmark, lw=lw, ls=ls, label=benchmark.name, color=colors[0])
- alpha = .25 if grayscale else 1
+ alpha = 0.25 if grayscale else 1
if isinstance(returns, _pd.Series):
ax.plot(returns, lw=lw, label=returns.name, color=colors[1], alpha=alpha)
elif isinstance(returns, _pd.DataFrame):
# color_dict = {col: colors[i+1] for i, col in enumerate(returns.columns)}
for i, col in enumerate(returns.columns):
- ax.plot(returns[col], lw=lw, label=col, alpha=alpha, color=colors[i+1])
+ ax.plot(returns[col], lw=lw, label=col, alpha=alpha, color=colors[i + 1])
if fill:
if isinstance(returns, _pd.Series):
- ax.fill_between(returns.index, 0, returns, color=colors[1], alpha=.25)
+ ax.fill_between(returns.index, 0, returns, color=colors[1], alpha=0.25)
elif isinstance(returns, _pd.DataFrame):
for i, col in enumerate(returns.columns):
- ax.fill_between(returns[col].index, 0, returns[col], color=colors[i+1], alpha=.25)
+ ax.fill_between(
+ returns[col].index, 0, returns[col], color=colors[i + 1], alpha=0.25
+ )
# rotate and align the tick labels so they look better
fig.autofmt_xdate()
@@ -281,14 +350,11 @@ def plot_timeseries(returns, benchmark=None,
if hline is not None:
if not isinstance(hline, _pd.Series):
if grayscale:
- hlcolor = 'black'
- ax.axhline(hline, ls="--", lw=hlw, color=hlcolor,
- label=hllabel, zorder=2)
+ hlcolor = "black"
+ ax.axhline(hline, ls="--", lw=hlw, color=hlcolor, label=hllabel, zorder=2)
- ax.axhline(0, ls="-", lw=1,
- color='gray', zorder=1)
- ax.axhline(0, ls="--", lw=1,
- color='white' if grayscale else 'black', zorder=2)
+ ax.axhline(0, ls="-", lw=1, color="gray", zorder=1)
+ ax.axhline(0, ls="--", lw=1, color="white" if grayscale else "black", zorder=2)
# if isinstance(benchmark, _pd.Series) or hline is not None:
ax.legend(fontsize=11)
@@ -300,11 +366,12 @@ def plot_timeseries(returns, benchmark=None,
# ax.yaxis.set_major_formatter(_plt.FuncFormatter(
# lambda x, loc: "{:,}%".format(int(x*100))))
- ax.set_xlabel('')
+ ax.set_xlabel("")
if ylabel:
- ax.set_ylabel(ylabel, fontname=fontname,
- fontweight='bold', fontsize=12, color="black")
- ax.yaxis.set_label_coords(-.1, .5)
+ ax.set_ylabel(
+ ylabel, fontname=fontname, fontweight="bold", fontsize=12, color="black"
+ )
+ ax.yaxis.set_label_coords(-0.1, 0.5)
if benchmark is None and len(_pd.DataFrame(returns).columns) == 1:
ax.get_legend().remove()
@@ -336,11 +403,22 @@ def plot_timeseries(returns, benchmark=None,
return None
-def plot_histogram(returns, benchmark, resample="M", bins=20,
- fontname='Arial', grayscale=False,
- title="Returns", kde=True, figsize=(10, 6),
- ylabel=True, subtitle=True, compounded=True,
- savefig=None, show=True):
+def plot_histogram(
+ returns,
+ benchmark,
+ resample="M",
+ bins=20,
+ fontname="Arial",
+ grayscale=False,
+ title="Returns",
+ kde=True,
+ figsize=(10, 6),
+ ylabel=True,
+ subtitle=True,
+ compounded=True,
+ savefig=None,
+ show=True,
+):
# colors = ['#348dc1', '#003366', 'red']
# if grayscale:
@@ -350,31 +428,43 @@ def plot_histogram(returns, benchmark, resample="M", bins=20,
apply_fnc = _stats.comp if compounded else _np.sum
if benchmark is not None:
- benchmark = benchmark.fillna(0).resample(resample).apply(
- apply_fnc).resample(resample).last()
-
- returns = returns.fillna(0).resample(resample).apply(
- apply_fnc).resample(resample).last()
+ benchmark = (
+ benchmark.fillna(0)
+ .resample(resample)
+ .apply(apply_fnc)
+ .resample(resample)
+ .last()
+ )
+
+ returns = (
+ returns.fillna(0).resample(resample).apply(apply_fnc).resample(resample).last()
+ )
figsize = (0.995 * figsize[0], figsize[1])
fig, ax = _plt.subplots(figsize=figsize)
- ax.spines['top'].set_visible(False)
- ax.spines['right'].set_visible(False)
- ax.spines['bottom'].set_visible(False)
- ax.spines['left'].set_visible(False)
+ ax.spines["top"].set_visible(False)
+ ax.spines["right"].set_visible(False)
+ ax.spines["bottom"].set_visible(False)
+ ax.spines["left"].set_visible(False)
- fig.suptitle(title, y=.94, fontweight="bold", fontname=fontname,
- fontsize=14, color="black")
+ fig.suptitle(
+ title, y=0.94, fontweight="bold", fontname=fontname, fontsize=14, color="black"
+ )
if subtitle:
- ax.set_title("%s - %s \n" % (
- returns.index.date[:1][0].strftime('%Y'),
- returns.index.date[-1:][0].strftime('%Y')
- ), fontsize=12, color='gray')
-
- fig.set_facecolor('white')
- ax.set_facecolor('white')
+ ax.set_title(
+ "%s - %s \n"
+ % (
+ returns.index.date[:1][0].strftime("%Y"),
+ returns.index.date[-1:][0].strftime("%Y"),
+ ),
+ fontsize=12,
+ color="gray",
+ )
+
+ fig.set_facecolor("white")
+ ax.set_facecolor("white")
if isinstance(returns, _pd.DataFrame) and len(returns.columns) == 1:
returns = returns[returns.columns[0]]
@@ -382,65 +472,102 @@ def plot_histogram(returns, benchmark, resample="M", bins=20,
pallete = colors[1:2] if benchmark is None else colors[:2]
alpha = 0.7
if isinstance(returns, _pd.DataFrame):
- pallete = colors[1:len(returns.columns)+1] if benchmark is None else colors[:len(returns.columns)+1]
+ pallete = (
+ colors[1 : len(returns.columns) + 1]
+ if benchmark is None
+ else colors[: len(returns.columns) + 1]
+ )
if len(returns.columns) > 1:
alpha = 0.5
if benchmark is not None:
if isinstance(returns, _pd.Series):
- combined_returns = benchmark.to_frame().join(returns.to_frame()) \
- .stack().reset_index() \
- .rename(columns={'level_1': '', 0: 'Returns'})
+ combined_returns = (
+ benchmark.to_frame()
+ .join(returns.to_frame())
+ .stack()
+ .reset_index()
+ .rename(columns={"level_1": "", 0: "Returns"})
+ )
elif isinstance(returns, _pd.DataFrame):
- combined_returns = benchmark.to_frame().join(returns) \
- .stack().reset_index() \
- .rename(columns={'level_1': '', 0: 'Returns'})
- x = _sns.histplot(data=combined_returns, x='Returns',
- bins=bins, alpha=alpha, kde=kde,
- stat="density", hue='',
- palette=pallete,
- ax=ax)
+ combined_returns = (
+ benchmark.to_frame()
+ .join(returns)
+ .stack()
+ .reset_index()
+ .rename(columns={"level_1": "", 0: "Returns"})
+ )
+ x = _sns.histplot(
+ data=combined_returns,
+ x="Returns",
+ bins=bins,
+ alpha=alpha,
+ kde=kde,
+ stat="density",
+ hue="",
+ palette=pallete,
+ ax=ax,
+ )
else:
if isinstance(returns, _pd.Series):
combined_returns = returns.copy()
if kde:
- _sns.kdeplot(data=combined_returns, color='black', ax=ax)
- x = _sns.histplot(data=combined_returns, bins=bins,
- alpha=alpha,
- kde=False,
- stat="density",
- color=colors[1],
- ax=ax)
+ _sns.kdeplot(data=combined_returns, color="black", ax=ax)
+ x = _sns.histplot(
+ data=combined_returns,
+ bins=bins,
+ alpha=alpha,
+ kde=False,
+ stat="density",
+ color=colors[1],
+ ax=ax,
+ )
elif isinstance(returns, _pd.DataFrame):
- combined_returns = returns.stack().reset_index() \
- .rename(columns={'level_1': '', 0: 'Returns'})
+ combined_returns = (
+ returns.stack()
+ .reset_index()
+ .rename(columns={"level_1": "", 0: "Returns"})
+ )
# _sns.kdeplot(data=combined_returns, color='black', ax=ax)
- x = _sns.histplot(data=combined_returns, x='Returns',
- bins=bins, alpha=alpha, kde=kde,
- stat="density", hue='',
- palette=pallete,
- ax=ax)
+ x = _sns.histplot(
+ data=combined_returns,
+ x="Returns",
+ bins=bins,
+ alpha=alpha,
+ kde=kde,
+ stat="density",
+ hue="",
+ palette=pallete,
+ ax=ax,
+ )
# Why do we need average?
if isinstance(combined_returns, _pd.Series) or len(combined_returns.columns) == 1:
- ax.axvline(combined_returns.mean(), ls="--", lw=1.5,
- zorder=2, label="Average", color="red")
-
+ ax.axvline(
+ combined_returns.mean(),
+ ls="--",
+ lw=1.5,
+ zorder=2,
+ label="Average",
+ color="red",
+ )
# _plt.setp(x.get_legend().get_texts(), fontsize=11)
- ax.xaxis.set_major_formatter(_plt.FuncFormatter(
- lambda x, loc: "{:,}%".format(int(x*100))))
+ ax.xaxis.set_major_formatter(
+ _plt.FuncFormatter(lambda x, loc: "{:,}%".format(int(x * 100)))
+ )
# Removed static lines for clarity
# ax.axhline(0.01, lw=1, color="#000000", zorder=2)
# ax.axvline(0, lw=1, color="#000000", zorder=2)
- ax.set_xlabel('')
- ax.set_ylabel("Occurrences", fontname=fontname,
- fontweight='bold', fontsize=12, color="black")
- ax.yaxis.set_label_coords(-.1, .5)
+ ax.set_xlabel("")
+ ax.set_ylabel(
+ "Occurrences", fontname=fontname, fontweight="bold", fontsize=12, color="black"
+ )
+ ax.yaxis.set_label_coords(-0.1, 0.5)
# fig.autofmt_xdate()
@@ -471,20 +598,32 @@ def plot_histogram(returns, benchmark, resample="M", bins=20,
return None
-def plot_rolling_stats(returns, benchmark=None, title="",
- returns_label="Strategy",
- hline=None, hlw=None, hlcolor="red", hllabel="",
- lw=1.5, figsize=(10, 6), ylabel="",
- grayscale=False, fontname="Arial", subtitle=True,
- savefig=None, show=True):
+def plot_rolling_stats(
+ returns,
+ benchmark=None,
+ title="",
+ returns_label="Strategy",
+ hline=None,
+ hlw=None,
+ hlcolor="red",
+ hllabel="",
+ lw=1.5,
+ figsize=(10, 6),
+ ylabel="",
+ grayscale=False,
+ fontname="Arial",
+ subtitle=True,
+ savefig=None,
+ show=True,
+):
colors, _, _ = _get_colors(grayscale)
fig, ax = _plt.subplots(figsize=figsize)
- ax.spines['top'].set_visible(False)
- ax.spines['right'].set_visible(False)
- ax.spines['bottom'].set_visible(False)
- ax.spines['left'].set_visible(False)
+ ax.spines["top"].set_visible(False)
+ ax.spines["right"].set_visible(False)
+ ax.spines["bottom"].set_visible(False)
+ ax.spines["left"].set_visible(False)
if isinstance(returns, _pd.DataFrame):
returns_label = list(returns.columns)
@@ -492,59 +631,70 @@ def plot_rolling_stats(returns, benchmark=None, title="",
if isinstance(returns, _pd.Series):
df = _pd.DataFrame(index=returns.index, data={returns_label: returns})
elif isinstance(returns, _pd.DataFrame):
- df = _pd.DataFrame(index=returns.index, data={col: returns[col] for col in returns.columns})
+ df = _pd.DataFrame(
+ index=returns.index, data={col: returns[col] for col in returns.columns}
+ )
if isinstance(benchmark, _pd.Series):
- df['Benchmark'] = benchmark[benchmark.index.isin(returns.index)]
+ df["Benchmark"] = benchmark[benchmark.index.isin(returns.index)]
if isinstance(returns, _pd.Series):
- df = df[['Benchmark', returns_label]].dropna()
- ax.plot(df[returns_label].dropna(), lw=lw,
- label=returns.name, color=colors[1])
+ df = df[["Benchmark", returns_label]].dropna()
+ ax.plot(
+ df[returns_label].dropna(), lw=lw, label=returns.name, color=colors[1]
+ )
elif isinstance(returns, _pd.DataFrame):
- col_names = ['Benchmark', returns_label]
+ col_names = ["Benchmark", returns_label]
df = df[list(_pd.core.common.flatten(col_names))].dropna()
for i, col in enumerate(returns_label):
- ax.plot(df[col], lw=lw, label=col, color=colors[i+1])
- ax.plot(df['Benchmark'], lw=lw, label=benchmark.name,
- color=colors[0], alpha=.8)
+ ax.plot(df[col], lw=lw, label=col, color=colors[i + 1])
+ ax.plot(
+ df["Benchmark"], lw=lw, label=benchmark.name, color=colors[0], alpha=0.8
+ )
else:
if isinstance(returns, _pd.Series):
df = df[[returns_label]].dropna()
- ax.plot(df[returns_label].dropna(), lw=lw,
- label=returns.name, color=colors[1])
+ ax.plot(
+ df[returns_label].dropna(), lw=lw, label=returns.name, color=colors[1]
+ )
elif isinstance(returns, _pd.DataFrame):
df = df[returns_label].dropna()
for i, col in enumerate(returns_label):
- ax.plot(df[col], lw=lw, label=col, color=colors[i+1])
+ ax.plot(df[col], lw=lw, label=col, color=colors[i + 1])
# rotate and align the tick labels so they look better
fig.autofmt_xdate()
# use a more precise date string for the x axis locations in the toolbar
# ax.fmt_xdata = _mdates.DateFormatter('%Y-%m-%d')\
- fig.suptitle(title, y=.94, fontweight="bold", fontname=fontname,
- fontsize=14, color="black")
+ fig.suptitle(
+ title, y=0.94, fontweight="bold", fontname=fontname, fontsize=14, color="black"
+ )
if subtitle:
- ax.set_title("%s - %s \n" % (
- df.index.date[:1][0].strftime('%e %b \'%y'),
- df.index.date[-1:][0].strftime('%e %b \'%y')
- ), fontsize=12, color='gray')
+ ax.set_title(
+ "%s - %s \n"
+ % (
+ df.index.date[:1][0].strftime("%e %b '%y"),
+ df.index.date[-1:][0].strftime("%e %b '%y"),
+ ),
+ fontsize=12,
+ color="gray",
+ )
if hline is not None:
if not isinstance(hline, _pd.Series):
if grayscale:
- hlcolor = 'black'
- ax.axhline(hline, ls="--", lw=hlw, color=hlcolor,
- label=hllabel, zorder=2)
+ hlcolor = "black"
+ ax.axhline(hline, ls="--", lw=hlw, color=hlcolor, label=hllabel, zorder=2)
ax.axhline(0, ls="--", lw=1, color="#000000", zorder=2)
if ylabel:
- ax.set_ylabel(ylabel, fontname=fontname,
- fontweight='bold', fontsize=12, color="black")
- ax.yaxis.set_label_coords(-.1, .5)
+ ax.set_ylabel(
+ ylabel, fontname=fontname, fontweight="bold", fontsize=12, color="black"
+ )
+ ax.yaxis.set_label_coords(-0.1, 0.5)
- ax.yaxis.set_major_formatter(_FormatStrFormatter('%.2f'))
+ ax.yaxis.set_major_formatter(_FormatStrFormatter("%.2f"))
ax.legend(fontsize=11)
@@ -577,37 +727,59 @@ def plot_rolling_stats(returns, benchmark=None, title="",
return None
-def plot_rolling_beta(returns, benchmark,
- window1=126, window1_label="",
- window2=None, window2_label="",
- title="", hlcolor="red", figsize=(10, 6),
- grayscale=False, fontname="Arial", lw=1.5,
- ylabel=True, subtitle=True, savefig=None, show=True):
+def plot_rolling_beta(
+ returns,
+ benchmark,
+ window1=126,
+ window1_label="",
+ window2=None,
+ window2_label="",
+ title="",
+ hlcolor="red",
+ figsize=(10, 6),
+ grayscale=False,
+ fontname="Arial",
+ lw=1.5,
+ ylabel=True,
+ subtitle=True,
+ savefig=None,
+ show=True,
+):
colors, _, _ = _get_colors(grayscale)
fig, ax = _plt.subplots(figsize=figsize)
- ax.spines['top'].set_visible(False)
- ax.spines['right'].set_visible(False)
- ax.spines['bottom'].set_visible(False)
- ax.spines['left'].set_visible(False)
+ ax.spines["top"].set_visible(False)
+ ax.spines["right"].set_visible(False)
+ ax.spines["bottom"].set_visible(False)
+ ax.spines["left"].set_visible(False)
- fig.suptitle(title, y=.94, fontweight="bold", fontname=fontname,
- fontsize=14, color="black")
+ fig.suptitle(
+ title, y=0.94, fontweight="bold", fontname=fontname, fontsize=14, color="black"
+ )
if subtitle:
- ax.set_title("%s - %s \n" % (
- returns.index.date[:1][0].strftime('%e %b \'%y'),
- returns.index.date[-1:][0].strftime('%e %b \'%y')
- ), fontsize=12, color='gray')
+ ax.set_title(
+ "%s - %s \n"
+ % (
+ returns.index.date[:1][0].strftime("%e %b '%y"),
+ returns.index.date[-1:][0].strftime("%e %b '%y"),
+ ),
+ fontsize=12,
+ color="gray",
+ )
i = 1
if isinstance(returns, _pd.Series):
- beta = _stats.rolling_greeks(returns, benchmark, window1)['beta'].fillna(0)
+ beta = _stats.rolling_greeks(returns, benchmark, window1)["beta"].fillna(0)
ax.plot(beta, lw=lw, label=window1_label, color=colors[1])
elif isinstance(returns, _pd.DataFrame):
- beta = ({col: _stats.rolling_greeks(returns[col], benchmark, window1)['beta'].fillna(0)
- for col in returns.columns})
+ beta = {
+ col: _stats.rolling_greeks(returns[col], benchmark, window1)["beta"].fillna(
+ 0
+ )
+ for col in returns.columns
+ }
for name, b in beta.items():
ax.plot(b, lw=lw, label=name + " " + f"({window1_label})", color=colors[i])
i += 1
@@ -616,38 +788,60 @@ def plot_rolling_beta(returns, benchmark,
if window2:
lw = lw - 0.5
if isinstance(returns, _pd.Series):
- ax.plot(_stats.rolling_greeks(returns, benchmark, window2)['beta'],
- lw=lw, label=window2_label, color="gray", alpha=0.8)
+ ax.plot(
+ _stats.rolling_greeks(returns, benchmark, window2)["beta"],
+ lw=lw,
+ label=window2_label,
+ color="gray",
+ alpha=0.8,
+ )
elif isinstance(returns, _pd.DataFrame):
- betas_w2 = ({col: _stats.rolling_greeks(returns[col], benchmark, window2)['beta']
- for col in returns.columns})
+ betas_w2 = {
+ col: _stats.rolling_greeks(returns[col], benchmark, window2)["beta"]
+ for col in returns.columns
+ }
for name, beta_w2 in betas_w2.items():
- ax.plot(beta_w2, lw=lw, ls='--', label=name + " " + f"({window2_label})", alpha=0.5, color=colors[i])
+ ax.plot(
+ beta_w2,
+ lw=lw,
+ ls="--",
+ label=name + " " + f"({window2_label})",
+ alpha=0.5,
+ color=colors[i],
+ )
i += 1
- beta_min = beta.min() if isinstance(returns, _pd.Series) else min([b.min() for b in beta.values()])
- beta_max = beta.max() if isinstance(returns, _pd.Series) else max([b.max() for b in beta.values()])
- mmin = min([-100, int(beta_min*100)])
- mmax = max([100, int(beta_max*100)])
- step = 50 if (mmax-mmin) >= 200 else 100
+ beta_min = (
+ beta.min()
+ if isinstance(returns, _pd.Series)
+ else min([b.min() for b in beta.values()])
+ )
+ beta_max = (
+ beta.max()
+ if isinstance(returns, _pd.Series)
+ else max([b.max() for b in beta.values()])
+ )
+ mmin = min([-100, int(beta_min * 100)])
+ mmax = max([100, int(beta_max * 100)])
+ step = 50 if (mmax - mmin) >= 200 else 100
ax.set_yticks([x / 100 for x in list(range(mmin, mmax, step))])
if isinstance(returns, _pd.Series):
- hlcolor = 'black' if grayscale else hlcolor
- ax.axhline(beta.mean(), ls="--", lw=1.5,
- color=hlcolor, zorder=2)
+ hlcolor = "black" if grayscale else hlcolor
+ ax.axhline(beta.mean(), ls="--", lw=1.5, color=hlcolor, zorder=2)
ax.axhline(0, ls="--", lw=1, color="#000000", zorder=2)
fig.autofmt_xdate()
# use a more precise date string for the x axis locations in the toolbar
- ax.fmt_xdata = _mdates.DateFormatter('%Y-%m-%d')
+ ax.fmt_xdata = _mdates.DateFormatter("%Y-%m-%d")
if ylabel:
- ax.set_ylabel("Beta", fontname=fontname,
- fontweight='bold', fontsize=12, color="black")
- ax.yaxis.set_label_coords(-.1, .5)
+ ax.set_ylabel(
+ "Beta", fontname=fontname, fontweight="bold", fontsize=12, color="black"
+ )
+ ax.yaxis.set_label_coords(-0.1, 0.5)
ax.legend(fontsize=11)
if benchmark is None and len(_pd.DataFrame(returns).columns) == 1:
@@ -680,58 +874,87 @@ def plot_rolling_beta(returns, benchmark,
return None
-def plot_longest_drawdowns(returns, periods=5, lw=1.5,
- fontname='Arial', grayscale=False, title=None,
- log_scale=False, figsize=(10, 6), ylabel=True,
- subtitle=True, compounded=True,
- savefig=None, show=True):
-
- colors = ['#348dc1', '#003366', 'red']
+def plot_longest_drawdowns(
+ returns,
+ periods=5,
+ lw=1.5,
+ fontname="Arial",
+ grayscale=False,
+ title=None,
+ log_scale=False,
+ figsize=(10, 6),
+ ylabel=True,
+ subtitle=True,
+ compounded=True,
+ savefig=None,
+ show=True,
+):
+
+ colors = ["#348dc1", "#003366", "red"]
if grayscale:
- colors = ['#000000'] * 3
+ colors = ["#000000"] * 3
dd = _stats.to_drawdown_series(returns.fillna(0))
dddf = _stats.drawdown_details(dd)
- longest_dd = dddf.sort_values(
- by='days', ascending=False, kind='mergesort')[:periods]
+ longest_dd = dddf.sort_values(by="days", ascending=False, kind="mergesort")[
+ :periods
+ ]
fig, ax = _plt.subplots(figsize=figsize)
- ax.spines['top'].set_visible(False)
- ax.spines['right'].set_visible(False)
- ax.spines['bottom'].set_visible(False)
- ax.spines['left'].set_visible(False)
-
- fig.suptitle(f"{title} - Worst %.0f Drawdown Periods" %
- periods, y=.94, fontweight="bold", fontname=fontname,
- fontsize=14, color="black")
+ ax.spines["top"].set_visible(False)
+ ax.spines["right"].set_visible(False)
+ ax.spines["bottom"].set_visible(False)
+ ax.spines["left"].set_visible(False)
+
+ fig.suptitle(
+ f"{title} - Worst %.0f Drawdown Periods" % periods,
+ y=0.94,
+ fontweight="bold",
+ fontname=fontname,
+ fontsize=14,
+ color="black",
+ )
if subtitle:
- ax.set_title("%s - %s \n" % (
- returns.index.date[:1][0].strftime('%e %b \'%y'),
- returns.index.date[-1:][0].strftime('%e %b \'%y')
- ), fontsize=12, color='gray')
-
- fig.set_facecolor('white')
- ax.set_facecolor('white')
+ ax.set_title(
+ "%s - %s \n"
+ % (
+ returns.index.date[:1][0].strftime("%e %b '%y"),
+ returns.index.date[-1:][0].strftime("%e %b '%y"),
+ ),
+ fontsize=12,
+ color="gray",
+ )
+
+ fig.set_facecolor("white")
+ ax.set_facecolor("white")
series = _stats.compsum(returns) if compounded else returns.cumsum()
ax.plot(series, lw=lw, label="Backtest", color=colors[0])
- highlight = 'black' if grayscale else 'red'
+ highlight = "black" if grayscale else "red"
for _, row in longest_dd.iterrows():
- ax.axvspan(*_mdates.datestr2num([str(row['start']), str(row['end'])]),
- color=highlight, alpha=.1)
+ ax.axvspan(
+ *_mdates.datestr2num([str(row["start"]), str(row["end"])]),
+ color=highlight,
+ alpha=0.1,
+ )
# rotate and align the tick labels so they look better
fig.autofmt_xdate()
# use a more precise date string for the x axis locations in the toolbar
- ax.fmt_xdata = _mdates.DateFormatter('%Y-%m-%d')
+ ax.fmt_xdata = _mdates.DateFormatter("%Y-%m-%d")
ax.axhline(0, ls="--", lw=1, color="#000000", zorder=2)
_plt.yscale("symlog" if log_scale else "linear")
if ylabel:
- ax.set_ylabel("Cumulative Returns", fontname=fontname,
- fontweight='bold', fontsize=12, color="black")
- ax.yaxis.set_label_coords(-.1, .5)
+ ax.set_ylabel(
+ "Cumulative Returns",
+ fontname=fontname,
+ fontweight="bold",
+ fontsize=12,
+ color="black",
+ )
+ ax.yaxis.set_label_coords(-0.1, 0.5)
ax.yaxis.set_major_formatter(_FuncFormatter(format_pct_axis))
# ax.yaxis.set_major_formatter(_plt.FuncFormatter(
@@ -766,76 +989,90 @@ def plot_longest_drawdowns(returns, periods=5, lw=1.5,
return None
-def plot_distribution(returns, figsize=(10, 6),
- fontname='Arial', grayscale=False, ylabel=True,
- subtitle=True, compounded=True, title=None,
- savefig=None, show=True):
+def plot_distribution(
+ returns,
+ figsize=(10, 6),
+ fontname="Arial",
+ grayscale=False,
+ ylabel=True,
+ subtitle=True,
+ compounded=True,
+ title=None,
+ savefig=None,
+ show=True,
+):
colors = _FLATUI_COLORS
if grayscale:
- colors = ['#f9f9f9', '#dddddd', '#bbbbbb', '#999999', '#808080']
+ colors = ["#f9f9f9", "#dddddd", "#bbbbbb", "#999999", "#808080"]
# colors, ls, alpha = _get_colors(grayscale)
port = _pd.DataFrame(returns.fillna(0))
- port.columns = ['Daily']
+ port.columns = ["Daily"]
apply_fnc = _stats.comp if compounded else _np.sum
- port['Weekly'] = port['Daily'].resample(
- 'W-MON').apply(apply_fnc)
- port['Weekly'].ffill(inplace=True)
+ port["Weekly"] = port["Daily"].resample("W-MON").apply(apply_fnc)
+ port["Weekly"].ffill(inplace=True)
- port['Monthly'] = port['Daily'].resample(
- 'M').apply(apply_fnc)
- port['Monthly'].ffill(inplace=True)
+ port["Monthly"] = port["Daily"].resample("M").apply(apply_fnc)
+ port["Monthly"].ffill(inplace=True)
- port['Quarterly'] = port['Daily'].resample(
- 'Q').apply(apply_fnc)
- port['Quarterly'].ffill(inplace=True)
+ port["Quarterly"] = port["Daily"].resample("Q").apply(apply_fnc)
+ port["Quarterly"].ffill(inplace=True)
- port['Yearly'] = port['Daily'].resample(
- 'A').apply(apply_fnc)
- port['Yearly'].ffill(inplace=True)
+ port["Yearly"] = port["Daily"].resample("A").apply(apply_fnc)
+ port["Yearly"].ffill(inplace=True)
fig, ax = _plt.subplots(figsize=figsize)
- ax.spines['top'].set_visible(False)
- ax.spines['right'].set_visible(False)
- ax.spines['bottom'].set_visible(False)
- ax.spines['left'].set_visible(False)
+ ax.spines["top"].set_visible(False)
+ ax.spines["right"].set_visible(False)
+ ax.spines["bottom"].set_visible(False)
+ ax.spines["left"].set_visible(False)
if title:
title = f"{title} - Return Quantiles"
else:
title = "Return Quantiles"
- fig.suptitle(title, y=.94,
- fontweight="bold", fontname=fontname,
- fontsize=14, color="black")
+ fig.suptitle(
+ title, y=0.94, fontweight="bold", fontname=fontname, fontsize=14, color="black"
+ )
if subtitle:
- ax.set_title("%s - %s \n" % (
- returns.index.date[:1][0].strftime('%e %b \'%y'),
- returns.index.date[-1:][0].strftime('%e %b \'%y')
- ), fontsize=12, color='gray')
-
- fig.set_facecolor('white')
- ax.set_facecolor('white')
-
- _sns.boxplot(data=port, ax=ax,
- palette={
- 'Daily': colors[0],
- 'Weekly': colors[1],
- 'Monthly': colors[2],
- 'Quarterly': colors[3],
- 'Yearly': colors[4]
- })
-
- ax.yaxis.set_major_formatter(_plt.FuncFormatter(
- lambda x, loc: "{:,}%".format(int(x*100))))
+ ax.set_title(
+ "%s - %s \n"
+ % (
+ returns.index.date[:1][0].strftime("%e %b '%y"),
+ returns.index.date[-1:][0].strftime("%e %b '%y"),
+ ),
+ fontsize=12,
+ color="gray",
+ )
+
+ fig.set_facecolor("white")
+ ax.set_facecolor("white")
+
+ _sns.boxplot(
+ data=port,
+ ax=ax,
+ palette={
+ "Daily": colors[0],
+ "Weekly": colors[1],
+ "Monthly": colors[2],
+ "Quarterly": colors[3],
+ "Yearly": colors[4],
+ },
+ )
+
+ ax.yaxis.set_major_formatter(
+ _plt.FuncFormatter(lambda x, loc: "{:,}%".format(int(x * 100)))
+ )
if ylabel:
- ax.set_ylabel('Returns', fontname=fontname,
- fontweight='bold', fontsize=12, color="black")
- ax.yaxis.set_label_coords(-.1, .5)
+ ax.set_ylabel(
+ "Returns", fontname=fontname, fontweight="bold", fontsize=12, color="black"
+ )
+ ax.yaxis.set_label_coords(-0.1, 0.5)
fig.autofmt_xdate()
@@ -865,17 +1102,22 @@ def plot_distribution(returns, figsize=(10, 6),
return None
-def plot_table(tbl, columns=None, title="", title_loc="left",
- header=True,
- colWidths=None,
- rowLoc='right',
- colLoc='right',
- colLabels=None,
- edges='horizontal',
- orient='horizontal',
- figsize=(5.5, 6),
- savefig=None,
- show=False):
+def plot_table(
+ tbl,
+ columns=None,
+ title="",
+ title_loc="left",
+ header=True,
+ colWidths=None,
+ rowLoc="right",
+ colLoc="right",
+ colLabels=None,
+ edges="horizontal",
+ orient="horizontal",
+ figsize=(5.5, 6),
+ savefig=None,
+ show=False,
+):
if columns is not None:
try:
@@ -887,18 +1129,20 @@ def plot_table(tbl, columns=None, title="", title_loc="left",
ax = _plt.subplot(111, frame_on=False)
if title != "":
- ax.set_title(title, fontweight="bold",
- fontsize=14, color="black", loc=title_loc)
-
- the_table = ax.table(cellText=tbl.values,
- colWidths=colWidths,
- rowLoc=rowLoc,
- colLoc=colLoc,
- edges=edges,
- colLabels=(tbl.columns if header else colLabels),
- loc='center',
- zorder=2
- )
+ ax.set_title(
+ title, fontweight="bold", fontsize=14, color="black", loc=title_loc
+ )
+
+ the_table = ax.table(
+ cellText=tbl.values,
+ colWidths=colWidths,
+ rowLoc=rowLoc,
+ colLoc=colLoc,
+ edges=edges,
+ colLabels=(tbl.columns if header else colLabels),
+ loc="center",
+ zorder=2,
+ )
the_table.auto_set_font_size(False)
the_table.set_fontsize(12)
@@ -906,17 +1150,17 @@ def plot_table(tbl, columns=None, title="", title_loc="left",
for (row, col), cell in the_table.get_celld().items():
cell.set_height(0.08)
- cell.set_text_props(color='black')
- cell.set_edgecolor('#dddddd')
+ cell.set_text_props(color="black")
+ cell.set_edgecolor("#dddddd")
if row == 0 and header:
- cell.set_edgecolor('black')
- cell.set_facecolor('black')
+ cell.set_edgecolor("black")
+ cell.set_facecolor("black")
cell.set_linewidth(2)
- cell.set_text_props(weight='bold', color='black')
+ cell.set_text_props(weight="bold", color="black")
elif col == 0 and "vertical" in orient:
- cell.set_edgecolor('#dddddd')
+ cell.set_edgecolor("#dddddd")
cell.set_linewidth(1)
- cell.set_text_props(weight='bold', color='black')
+ cell.set_text_props(weight="bold", color="black")
elif row > 1:
cell.set_linewidth(1)
@@ -952,34 +1196,34 @@ def plot_table(tbl, columns=None, title="", title_loc="left",
def format_cur_axis(x, _):
if x >= 1e12:
- res = '$%1.1fT' % (x * 1e-12)
- return res.replace('.0T', 'T')
+ res = "$%1.1fT" % (x * 1e-12)
+ return res.replace(".0T", "T")
if x >= 1e9:
- res = '$%1.1fB' % (x * 1e-9)
- return res.replace('.0B', 'B')
+ res = "$%1.1fB" % (x * 1e-9)
+ return res.replace(".0B", "B")
if x >= 1e6:
- res = '$%1.1fM' % (x * 1e-6)
- return res.replace('.0M', 'M')
+ res = "$%1.1fM" % (x * 1e-6)
+ return res.replace(".0M", "M")
if x >= 1e3:
- res = '$%1.0fK' % (x * 1e-3)
- return res.replace('.0K', 'K')
- res = '$%1.0f' % x
- return res.replace('.0', '')
+ res = "$%1.0fK" % (x * 1e-3)
+ return res.replace(".0K", "K")
+ res = "$%1.0f" % x
+ return res.replace(".0", "")
def format_pct_axis(x, _):
x *= 100 # lambda x, loc: "{:,}%".format(int(x * 100))
if x >= 1e12:
- res = '%1.1fT%%' % (x * 1e-12)
- return res.replace('.0T%', 'T%')
+ res = "%1.1fT%%" % (x * 1e-12)
+ return res.replace(".0T%", "T%")
if x >= 1e9:
- res = '%1.1fB%%' % (x * 1e-9)
- return res.replace('.0B%', 'B%')
+ res = "%1.1fB%%" % (x * 1e-9)
+ return res.replace(".0B%", "B%")
if x >= 1e6:
- res = '%1.1fM%%' % (x * 1e-6)
- return res.replace('.0M%', 'M%')
+ res = "%1.1fM%%" % (x * 1e-6)
+ return res.replace(".0M%", "M%")
if x >= 1e3:
- res = '%1.1fK%%' % (x * 1e-3)
- return res.replace('.0K%', 'K%')
- res = '%1.0f%%' % x
- return res.replace('.0%', '%')
+ res = "%1.1fK%%" % (x * 1e-3)
+ return res.replace(".0K%", "K%")
+ res = "%1.0f%%" % x
+ return res.replace(".0%", "%")
diff --git a/quantstats/_plotting/wrappers.py b/quantstats/_plotting/wrappers.py
index 10687ec4..90f02f83 100644
--- a/quantstats/_plotting/wrappers.py
+++ b/quantstats/_plotting/wrappers.py
@@ -22,7 +22,7 @@
import matplotlib.pyplot as _plt
from matplotlib.ticker import (
StrMethodFormatter as _StrMethodFormatter,
- FuncFormatter as _FuncFormatter
+ FuncFormatter as _FuncFormatter,
)
import numpy as _np
@@ -31,19 +31,20 @@
import seaborn as _sns
from .. import (
- stats as _stats, utils as _utils,
+ stats as _stats,
+ utils as _utils,
)
from . import core as _core
-_FLATUI_COLORS = ["#fedd78", "#348dc1", "#af4b64",
- "#4fa487", "#9b59b6", "#808080"]
-_GRAYSCALE_COLORS = (len(_FLATUI_COLORS) * ['black']) + ['white']
+_FLATUI_COLORS = ["#fedd78", "#348dc1", "#af4b64", "#4fa487", "#9b59b6", "#808080"]
+_GRAYSCALE_COLORS = (len(_FLATUI_COLORS) * ["black"]) + ["white"]
_HAS_PLOTLY = False
try:
import plotly
+
_HAS_PLOTLY = True
except ImportError:
pass
@@ -55,14 +56,23 @@ def to_plotly(fig):
with warnings.catch_warnings():
warnings.simplefilter("ignore")
fig = plotly.tools.mpl_to_plotly(fig)
- return plotly.plotly.iplot(fig, filename='quantstats-plot',
- overwrite=True)
-
-
-def snapshot(returns, grayscale=False, figsize=(10, 8),
- title='Portfolio Summary', fontname='Arial', lw=1.5,
- mode="comp", subtitle=True, savefig=None, show=True,
- log_scale=False, **kwargs):
+ return plotly.plotly.iplot(fig, filename="quantstats-plot", overwrite=True)
+
+
+def snapshot(
+ returns,
+ grayscale=False,
+ figsize=(10, 8),
+ title="Portfolio Summary",
+ fontname="Arial",
+ lw=1.5,
+ mode="comp",
+ subtitle=True,
+ savefig=None,
+ show=True,
+ log_scale=False,
+ **kwargs,
+):
strategy_colname = kwargs.get("strategy_col", "Strategy")
@@ -84,52 +94,79 @@ def snapshot(returns, grayscale=False, figsize=(10, 8),
if figsize is None:
size = list(_plt.gcf().get_size_inches())
- figsize = (size[0], size[0]*.75)
+ figsize = (size[0], size[0] * 0.75)
- fig, axes = _plt.subplots(3, 1, sharex=True, figsize=figsize,
- gridspec_kw={'height_ratios': [3, 1, 1]})
+ fig, axes = _plt.subplots(
+ 3, 1, sharex=True, figsize=figsize, gridspec_kw={"height_ratios": [3, 1, 1]}
+ )
if multi_column:
_plt.figtext(
- 0, -.05,
+ 0,
+ -0.05,
" * When a multi-column DataFrame is passed, the mean of all columns will be used as returns.\n"
" To change this behavior, use a pandas Series or pass the column name in the `strategy_col` parameter.",
- ha="left", fontsize=11, color='black', alpha=0.6, linespacing=1.5)
+ ha="left",
+ fontsize=11,
+ color="black",
+ alpha=0.6,
+ linespacing=1.5,
+ )
for ax in axes:
- ax.spines['top'].set_visible(False)
- ax.spines['right'].set_visible(False)
- ax.spines['bottom'].set_visible(False)
- ax.spines['left'].set_visible(False)
+ ax.spines["top"].set_visible(False)
+ ax.spines["right"].set_visible(False)
+ ax.spines["bottom"].set_visible(False)
+ ax.spines["left"].set_visible(False)
- fig.suptitle(title, fontsize=14, y=.97,
- fontname=fontname, fontweight='bold', color='black')
+ fig.suptitle(
+ title, fontsize=14, y=0.97, fontname=fontname, fontweight="bold", color="black"
+ )
- fig.set_facecolor('white')
+ fig.set_facecolor("white")
if subtitle:
if isinstance(returns, _pd.Series):
- axes[0].set_title("%s - %s ; Sharpe: %.2f \n" % (
- returns.index.date[:1][0].strftime('%e %b \'%y'),
- returns.index.date[-1:][0].strftime('%e %b \'%y'),
- _stats.sharpe(returns)
- ), fontsize=12, color='gray')
+ axes[0].set_title(
+ "%s - %s ; Sharpe: %.2f \n"
+ % (
+ returns.index.date[:1][0].strftime("%e %b '%y"),
+ returns.index.date[-1:][0].strftime("%e %b '%y"),
+ _stats.sharpe(returns),
+ ),
+ fontsize=12,
+ color="gray",
+ )
elif isinstance(returns, _pd.DataFrame):
- axes[0].set_title("\n%s - %s ; " % (
- returns.index.date[:1][0].strftime('%e %b \'%y'),
- returns.index.date[-1:][0].strftime('%e %b \'%y')
- ), fontsize=12, color='gray')
-
- axes[0].set_ylabel('Cumulative Return', fontname=fontname,
- fontweight='bold', fontsize=12)
+ axes[0].set_title(
+ "\n%s - %s ; "
+ % (
+ returns.index.date[:1][0].strftime("%e %b '%y"),
+ returns.index.date[-1:][0].strftime("%e %b '%y"),
+ ),
+ fontsize=12,
+ color="gray",
+ )
+
+ axes[0].set_ylabel(
+ "Cumulative Return", fontname=fontname, fontweight="bold", fontsize=12
+ )
if isinstance(returns, _pd.Series):
- axes[0].plot(_stats.compsum(returns) * 100, color=colors[1],
- lw=1 if grayscale else lw, zorder=1)
+ axes[0].plot(
+ _stats.compsum(returns) * 100,
+ color=colors[1],
+ lw=1 if grayscale else lw,
+ zorder=1,
+ )
elif isinstance(returns, _pd.DataFrame):
for col in returns.columns:
- axes[0].plot(_stats.compsum(returns[col]) * 100, label=col,
- lw=1 if grayscale else lw, zorder=1)
- axes[0].axhline(0, color='silver', lw=1, zorder=0)
+ axes[0].plot(
+ _stats.compsum(returns[col]) * 100,
+ label=col,
+ lw=1 if grayscale else lw,
+ zorder=1,
+ )
+ axes[0].axhline(0, color="silver", lw=1, zorder=0)
axes[0].set_yscale("symlog" if log_scale else "linear")
# axes[0].legend(fontsize=12)
@@ -144,41 +181,47 @@ def snapshot(returns, grayscale=False, figsize=(10, 8),
ddmin_ticks = int(_utils._round_to_closest(ddmin_ticks, 5))
# ddmin_ticks = int(_utils._round_to_closest(ddmin, 5))
- axes[1].set_ylabel('Drawdown', fontname=fontname,
- fontweight='bold', fontsize=12)
+ axes[1].set_ylabel("Drawdown", fontname=fontname, fontweight="bold", fontsize=12)
axes[1].set_yticks(_np.arange(-ddmin, 0, step=ddmin_ticks))
if isinstance(dd, _pd.Series):
axes[1].plot(dd, color=colors[2], lw=1 if grayscale else lw, zorder=1)
elif isinstance(dd, _pd.DataFrame):
for col in dd.columns:
axes[1].plot(dd[col], label=col, lw=1 if grayscale else lw, zorder=1)
- axes[1].axhline(0, color='silver', lw=1, zorder=0)
+ axes[1].axhline(0, color="silver", lw=1, zorder=0)
if not grayscale:
if isinstance(dd, _pd.Series):
- axes[1].fill_between(dd.index, 0, dd, color=colors[2], alpha=.25)
+ axes[1].fill_between(dd.index, 0, dd, color=colors[2], alpha=0.25)
elif isinstance(dd, _pd.DataFrame):
for i, col in enumerate(dd.columns):
- axes[1].fill_between(dd[col].index, 0, dd[col], color=colors[i + 1], alpha=.25)
+ axes[1].fill_between(
+ dd[col].index, 0, dd[col], color=colors[i + 1], alpha=0.25
+ )
axes[1].set_yscale("symlog" if log_scale else "linear")
# axes[1].legend(fontsize=12)
- axes[2].set_ylabel('Daily Return', fontname=fontname,
- fontweight='bold', fontsize=12)
+ axes[2].set_ylabel(
+ "Daily Return", fontname=fontname, fontweight="bold", fontsize=12
+ )
if isinstance(returns, _pd.Series):
- axes[2].plot(returns * 100, color=colors[0], label=returns.name, lw=0.5, zorder=1)
+ axes[2].plot(
+ returns * 100, color=colors[0], label=returns.name, lw=0.5, zorder=1
+ )
elif isinstance(returns, _pd.DataFrame):
for i, col in enumerate(returns.columns):
- axes[2].plot(returns[col] * 100, color=colors[i], label=col, lw=0.5, zorder=1)
- axes[2].axhline(0, color='silver', lw=1, zorder=0)
- axes[2].axhline(0, color=colors[-1], linestyle='--', lw=1, zorder=2)
+ axes[2].plot(
+ returns[col] * 100, color=colors[i], label=col, lw=0.5, zorder=1
+ )
+ axes[2].axhline(0, color="silver", lw=1, zorder=0)
+ axes[2].axhline(0, color=colors[-1], linestyle="--", lw=1, zorder=2)
axes[2].set_yscale("symlog" if log_scale else "linear")
# axes[2].legend(fontsize=12)
retmax = _utils._round_to_closest(returns.max() * 100, 5)
retmin = _utils._round_to_closest(returns.min() * 100, 5)
- retdiff = (retmax - retmin)
+ retdiff = retmax - retmin
steps = 5
if retdiff > 50:
steps = retdiff / 5
@@ -188,9 +231,9 @@ def snapshot(returns, grayscale=False, figsize=(10, 8),
axes[2].set_yticks(_np.arange(retmin, retmax, step=steps))
for ax in axes:
- ax.set_facecolor('white')
- ax.yaxis.set_label_coords(-.1, .5)
- ax.yaxis.set_major_formatter(_StrMethodFormatter('{x:,.0f}%'))
+ ax.set_facecolor("white")
+ ax.yaxis.set_label_coords(-0.1, 0.5)
+ ax.yaxis.set_major_formatter(_StrMethodFormatter("{x:,.0f}%"))
_plt.subplots_adjust(hspace=0, bottom=0, top=1)
fig.autofmt_xdate()
@@ -221,58 +264,86 @@ def snapshot(returns, grayscale=False, figsize=(10, 8),
return None
-def earnings(returns, start_balance=1e5, mode="comp",
- grayscale=False, figsize=(10, 6),
- title='Portfolio Earnings',
- fontname='Arial', lw=1.5,
- subtitle=True, savefig=None, show=True):
+def earnings(
+ returns,
+ start_balance=1e5,
+ mode="comp",
+ grayscale=False,
+ figsize=(10, 6),
+ title="Portfolio Earnings",
+ fontname="Arial",
+ lw=1.5,
+ subtitle=True,
+ savefig=None,
+ show=True,
+):
colors = _GRAYSCALE_COLORS if grayscale else _FLATUI_COLORS
- alpha = .5 if grayscale else .8
+ alpha = 0.5 if grayscale else 0.8
returns = _utils.make_portfolio(returns, start_balance, mode)
if figsize is None:
size = list(_plt.gcf().get_size_inches())
- figsize = (size[0], size[0]*.55)
+ figsize = (size[0], size[0] * 0.55)
fig, ax = _plt.subplots(figsize=figsize)
- ax.spines['top'].set_visible(False)
- ax.spines['right'].set_visible(False)
- ax.spines['bottom'].set_visible(False)
- ax.spines['left'].set_visible(False)
+ ax.spines["top"].set_visible(False)
+ ax.spines["right"].set_visible(False)
+ ax.spines["bottom"].set_visible(False)
+ ax.spines["left"].set_visible(False)
- fig.suptitle(title, fontsize=14, y=.995,
- fontname=fontname, fontweight='bold', color='black')
+ fig.suptitle(
+ title, fontsize=14, y=0.995, fontname=fontname, fontweight="bold", color="black"
+ )
if subtitle:
- ax.set_title("\n%s - %s ; P&L: %s (%s) " % (
- returns.index.date[1:2][0].strftime('%e %b \'%y'),
- returns.index.date[-1:][0].strftime('%e %b \'%y'),
- _utils._score_str("${:,}".format(
- round(returns.values[-1]-returns.values[0], 2))),
- _utils._score_str("{:,}%".format(
- round((returns.values[-1]/returns.values[0]-1)*100, 2)))
- ), fontsize=12, color='gray')
+ ax.set_title(
+ "\n%s - %s ; P&L: %s (%s) "
+ % (
+ returns.index.date[1:2][0].strftime("%e %b '%y"),
+ returns.index.date[-1:][0].strftime("%e %b '%y"),
+ _utils._score_str(
+ "${:,}".format(round(returns.values[-1] - returns.values[0], 2))
+ ),
+ _utils._score_str(
+ "{:,}%".format(
+ round((returns.values[-1] / returns.values[0] - 1) * 100, 2)
+ )
+ ),
+ ),
+ fontsize=12,
+ color="gray",
+ )
mx = returns.max()
returns_max = returns[returns == mx]
ix = returns_max[~_np.isnan(returns_max)].index[0]
returns_max = _np.where(returns.index == ix, mx, _np.nan)
- ax.plot(returns.index, returns_max, marker='o', lw=0,
- alpha=alpha, markersize=12, color=colors[0])
- ax.plot(returns.index, returns, color=colors[1],
- lw=1 if grayscale else lw)
-
- ax.set_ylabel('Value of ${:,.0f}'.format(start_balance),
- fontname=fontname, fontweight='bold', fontsize=12)
+ ax.plot(
+ returns.index,
+ returns_max,
+ marker="o",
+ lw=0,
+ alpha=alpha,
+ markersize=12,
+ color=colors[0],
+ )
+ ax.plot(returns.index, returns, color=colors[1], lw=1 if grayscale else lw)
+
+ ax.set_ylabel(
+ "Value of ${:,.0f}".format(start_balance),
+ fontname=fontname,
+ fontweight="bold",
+ fontsize=12,
+ )
ax.yaxis.set_major_formatter(_FuncFormatter(_core.format_cur_axis))
- ax.yaxis.set_label_coords(-.1, .5)
+ ax.yaxis.set_label_coords(-0.1, 0.5)
- fig.set_facecolor('white')
- ax.set_facecolor('white')
+ fig.set_facecolor("white")
+ ax.set_facecolor("white")
fig.autofmt_xdate()
try:
@@ -301,93 +372,132 @@ def earnings(returns, start_balance=1e5, mode="comp",
return None
-def returns(returns, benchmark=None,
- grayscale=False, figsize=(10, 6),
- fontname='Arial', lw=1.5,
- match_volatility=False, compound=True, cumulative=True,
- resample=None, ylabel="Cumulative Returns",
- subtitle=True, savefig=None, show=True,
- prepare_returns=True):
-
- title = 'Cumulative Returns' if compound else 'Returns'
+def returns(
+ returns,
+ benchmark=None,
+ grayscale=False,
+ figsize=(10, 6),
+ fontname="Arial",
+ lw=1.5,
+ match_volatility=False,
+ compound=True,
+ cumulative=True,
+ resample=None,
+ ylabel="Cumulative Returns",
+ subtitle=True,
+ savefig=None,
+ show=True,
+ prepare_returns=True,
+):
+
+ title = "Cumulative Returns" if compound else "Returns"
if benchmark is not None:
if isinstance(benchmark, str):
- title += ' vs %s' % benchmark.upper()
+ title += " vs %s" % benchmark.upper()
else:
- title += ' vs Benchmark'
+ title += " vs Benchmark"
if match_volatility:
- title += ' (Volatility Matched)'
+ title += " (Volatility Matched)"
benchmark = _utils._prepare_benchmark(benchmark, returns.index)
if prepare_returns:
returns = _utils._prepare_returns(returns)
- fig = _core.plot_timeseries(returns, benchmark, title,
- ylabel=ylabel,
- match_volatility=match_volatility,
- log_scale=False,
- resample=resample,
- compound=compound,
- cumulative=cumulative,
- lw=lw,
- figsize=figsize,
- fontname=fontname,
- grayscale=grayscale,
- subtitle=subtitle,
- savefig=savefig, show=show)
+ fig = _core.plot_timeseries(
+ returns,
+ benchmark,
+ title,
+ ylabel=ylabel,
+ match_volatility=match_volatility,
+ log_scale=False,
+ resample=resample,
+ compound=compound,
+ cumulative=cumulative,
+ lw=lw,
+ figsize=figsize,
+ fontname=fontname,
+ grayscale=grayscale,
+ subtitle=subtitle,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def log_returns(returns, benchmark=None,
- grayscale=False, figsize=(10, 5),
- fontname='Arial', lw=1.5,
- match_volatility=False, compound=True, cumulative=True,
- resample=None, ylabel="Cumulative Returns",
- subtitle=True, savefig=None, show=True,
- prepare_returns=True):
-
- title = 'Cumulative Returns' if compound else 'Returns'
+def log_returns(
+ returns,
+ benchmark=None,
+ grayscale=False,
+ figsize=(10, 5),
+ fontname="Arial",
+ lw=1.5,
+ match_volatility=False,
+ compound=True,
+ cumulative=True,
+ resample=None,
+ ylabel="Cumulative Returns",
+ subtitle=True,
+ savefig=None,
+ show=True,
+ prepare_returns=True,
+):
+
+ title = "Cumulative Returns" if compound else "Returns"
if benchmark is not None:
if isinstance(benchmark, str):
- title += ' vs %s (Log Scaled' % benchmark.upper()
+ title += " vs %s (Log Scaled" % benchmark.upper()
else:
- title += ' vs Benchmark (Log Scaled'
+ title += " vs Benchmark (Log Scaled"
if match_volatility:
- title += ', Volatility Matched'
+ title += ", Volatility Matched"
else:
- title += ' (Log Scaled'
- title += ')'
+ title += " (Log Scaled"
+ title += ")"
if prepare_returns:
returns = _utils._prepare_returns(returns)
benchmark = _utils._prepare_benchmark(benchmark, returns.index)
- fig = _core.plot_timeseries(returns, benchmark, title,
- ylabel=ylabel,
- match_volatility=match_volatility,
- log_scale=True,
- resample=resample,
- compound=compound,
- cumulative=cumulative,
- lw=lw,
- figsize=figsize,
- fontname=fontname,
- grayscale=grayscale,
- subtitle=subtitle,
- savefig=savefig, show=show)
+ fig = _core.plot_timeseries(
+ returns,
+ benchmark,
+ title,
+ ylabel=ylabel,
+ match_volatility=match_volatility,
+ log_scale=True,
+ resample=resample,
+ compound=compound,
+ cumulative=cumulative,
+ lw=lw,
+ figsize=figsize,
+ fontname=fontname,
+ grayscale=grayscale,
+ subtitle=subtitle,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def daily_returns(returns, benchmark,
- grayscale=False, figsize=(10, 4),
- fontname='Arial', lw=0.5,
- log_scale=False, ylabel="Returns",
- subtitle=True, savefig=None, show=True,
- prepare_returns=True, active=False):
+def daily_returns(
+ returns,
+ benchmark,
+ grayscale=False,
+ figsize=(10, 4),
+ fontname="Arial",
+ lw=0.5,
+ log_scale=False,
+ ylabel="Returns",
+ subtitle=True,
+ savefig=None,
+ show=True,
+ prepare_returns=True,
+ active=False,
+):
if prepare_returns:
returns = _utils._prepare_returns(returns)
@@ -395,297 +505,454 @@ def daily_returns(returns, benchmark,
benchmark = _utils._prepare_returns(benchmark)
returns = returns - benchmark
- plot_title = 'Daily Active Returns' if active else 'Daily Returns'
-
- fig = _core.plot_timeseries(returns, None, plot_title,
- ylabel=ylabel,
- match_volatility=False,
- log_scale=log_scale,
- resample='D',
- compound=False,
- lw=lw,
- figsize=figsize,
- fontname=fontname,
- grayscale=grayscale,
- subtitle=subtitle,
- savefig=savefig, show=show)
+ plot_title = "Daily Active Returns" if active else "Daily Returns"
+
+ fig = _core.plot_timeseries(
+ returns,
+ None,
+ plot_title,
+ ylabel=ylabel,
+ match_volatility=False,
+ log_scale=log_scale,
+ resample="D",
+ compound=False,
+ lw=lw,
+ figsize=figsize,
+ fontname=fontname,
+ grayscale=grayscale,
+ subtitle=subtitle,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def yearly_returns(returns, benchmark=None,
- fontname='Arial', grayscale=False,
- hlw=1.5, hlcolor="red", hllabel="",
- match_volatility=False,
- log_scale=False, figsize=(10, 5), ylabel=True,
- subtitle=True, compounded=True,
- savefig=None, show=True,
- prepare_returns=True):
-
- title = 'EOY Returns'
+def yearly_returns(
+ returns,
+ benchmark=None,
+ fontname="Arial",
+ grayscale=False,
+ hlw=1.5,
+ hlcolor="red",
+ hllabel="",
+ match_volatility=False,
+ log_scale=False,
+ figsize=(10, 5),
+ ylabel=True,
+ subtitle=True,
+ compounded=True,
+ savefig=None,
+ show=True,
+ prepare_returns=True,
+):
+
+ title = "EOY Returns"
if benchmark is not None:
- title += ' vs Benchmark'
- benchmark = _utils._prepare_benchmark(
- benchmark, returns.index).resample('A').apply(
- _stats.comp).resample('A').last()
+ title += " vs Benchmark"
+ benchmark = (
+ _utils._prepare_benchmark(benchmark, returns.index)
+ .resample("A")
+ .apply(_stats.comp)
+ .resample("A")
+ .last()
+ )
if prepare_returns:
returns = _utils._prepare_returns(returns)
if compounded:
- returns = returns.resample('A').apply(_stats.comp)
+ returns = returns.resample("A").apply(_stats.comp)
else:
- returns = returns.resample('A').apply(_df.sum)
- returns = returns.resample('A').last()
-
- fig = _core.plot_returns_bars(returns, benchmark,
- fontname=fontname,
- hline=returns.mean(),
- hlw=hlw,
- hllabel=hllabel,
- hlcolor=hlcolor,
- match_volatility=match_volatility,
- log_scale=log_scale,
- resample=None,
- title=title,
- figsize=figsize,
- grayscale=grayscale,
- ylabel=ylabel,
- subtitle=subtitle,
- savefig=savefig, show=show)
+ returns = returns.resample("A").apply(_df.sum)
+ returns = returns.resample("A").last()
+
+ fig = _core.plot_returns_bars(
+ returns,
+ benchmark,
+ fontname=fontname,
+ hline=returns.mean(),
+ hlw=hlw,
+ hllabel=hllabel,
+ hlcolor=hlcolor,
+ match_volatility=match_volatility,
+ log_scale=log_scale,
+ resample=None,
+ title=title,
+ figsize=figsize,
+ grayscale=grayscale,
+ ylabel=ylabel,
+ subtitle=subtitle,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def distribution(returns, fontname='Arial', grayscale=False, ylabel=True,
- figsize=(10, 6), subtitle=True, compounded=True,
- savefig=None, show=True, title=None,
- prepare_returns=True):
+def distribution(
+ returns,
+ fontname="Arial",
+ grayscale=False,
+ ylabel=True,
+ figsize=(10, 6),
+ subtitle=True,
+ compounded=True,
+ savefig=None,
+ show=True,
+ title=None,
+ prepare_returns=True,
+):
if prepare_returns:
returns = _utils._prepare_returns(returns)
- fig = _core.plot_distribution(returns,
- fontname=fontname,
- grayscale=grayscale,
- figsize=figsize,
- ylabel=ylabel,
- subtitle=subtitle,
- title=title,
- compounded=compounded,
- savefig=savefig, show=show)
+ fig = _core.plot_distribution(
+ returns,
+ fontname=fontname,
+ grayscale=grayscale,
+ figsize=figsize,
+ ylabel=ylabel,
+ subtitle=subtitle,
+ title=title,
+ compounded=compounded,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def histogram(returns, benchmark=None, resample='M', fontname='Arial',
- grayscale=False, figsize=(10, 5), ylabel=True,
- subtitle=True, compounded=True, savefig=None, show=True,
- prepare_returns=True):
+def histogram(
+ returns,
+ benchmark=None,
+ resample="M",
+ fontname="Arial",
+ grayscale=False,
+ figsize=(10, 5),
+ ylabel=True,
+ subtitle=True,
+ compounded=True,
+ savefig=None,
+ show=True,
+ prepare_returns=True,
+):
if prepare_returns:
returns = _utils._prepare_returns(returns)
if benchmark is not None:
benchmark = _utils._prepare_returns(benchmark)
- if resample == 'W':
+ if resample == "W":
title = "Weekly "
- elif resample == 'M':
+ elif resample == "M":
title = "Monthly "
- elif resample == 'Q':
+ elif resample == "Q":
title = "Quarterly "
- elif resample == 'A':
+ elif resample == "A":
title = "Annual "
else:
title = ""
- return _core.plot_histogram(returns,
- benchmark,
- resample=resample,
- grayscale=grayscale,
- fontname=fontname,
- title="Distribution of %sReturns" % title,
- figsize=figsize,
- ylabel=ylabel,
- subtitle=subtitle,
- compounded=compounded,
- savefig=savefig, show=show)
-
-
-def drawdown(returns, grayscale=False, figsize=(10, 5),
- fontname='Arial', lw=1, log_scale=False,
- match_volatility=False, compound=False, ylabel="Drawdown",
- resample=None, subtitle=True, savefig=None, show=True):
+ return _core.plot_histogram(
+ returns,
+ benchmark,
+ resample=resample,
+ grayscale=grayscale,
+ fontname=fontname,
+ title="Distribution of %sReturns" % title,
+ figsize=figsize,
+ ylabel=ylabel,
+ subtitle=subtitle,
+ compounded=compounded,
+ savefig=savefig,
+ show=show,
+ )
+
+
+def drawdown(
+ returns,
+ grayscale=False,
+ figsize=(10, 5),
+ fontname="Arial",
+ lw=1,
+ log_scale=False,
+ match_volatility=False,
+ compound=False,
+ ylabel="Drawdown",
+ resample=None,
+ subtitle=True,
+ savefig=None,
+ show=True,
+):
dd = _stats.to_drawdown_series(returns)
- fig = _core.plot_timeseries(dd, title='Underwater Plot',
- hline=dd.mean(), hlw=2, hllabel="Average",
- returns_label="Drawdown",
- compound=compound, match_volatility=match_volatility,
- log_scale=log_scale, resample=resample,
- fill=True, lw=lw, figsize=figsize,
- ylabel=ylabel,
- fontname=fontname, grayscale=grayscale,
- subtitle=subtitle,
- savefig=savefig, show=show)
+ fig = _core.plot_timeseries(
+ dd,
+ title="Underwater Plot",
+ hline=dd.mean(),
+ hlw=2,
+ hllabel="Average",
+ returns_label="Drawdown",
+ compound=compound,
+ match_volatility=match_volatility,
+ log_scale=log_scale,
+ resample=resample,
+ fill=True,
+ lw=lw,
+ figsize=figsize,
+ ylabel=ylabel,
+ fontname=fontname,
+ grayscale=grayscale,
+ subtitle=subtitle,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def drawdowns_periods(returns, periods=5, lw=1.5, log_scale=False,
- fontname='Arial', grayscale=False, title=None, figsize=(10, 5),
- ylabel=True, subtitle=True, compounded=True,
- savefig=None, show=True,
- prepare_returns=True):
+def drawdowns_periods(
+ returns,
+ periods=5,
+ lw=1.5,
+ log_scale=False,
+ fontname="Arial",
+ grayscale=False,
+ title=None,
+ figsize=(10, 5),
+ ylabel=True,
+ subtitle=True,
+ compounded=True,
+ savefig=None,
+ show=True,
+ prepare_returns=True,
+):
if prepare_returns:
returns = _utils._prepare_returns(returns)
- fig = _core.plot_longest_drawdowns(returns,
- periods=periods,
- lw=lw,
- log_scale=log_scale,
- fontname=fontname,
- grayscale=grayscale,
- title=title,
- figsize=figsize,
- ylabel=ylabel,
- subtitle=subtitle,
- compounded=compounded,
- savefig=savefig, show=show)
+ fig = _core.plot_longest_drawdowns(
+ returns,
+ periods=periods,
+ lw=lw,
+ log_scale=log_scale,
+ fontname=fontname,
+ grayscale=grayscale,
+ title=title,
+ figsize=figsize,
+ ylabel=ylabel,
+ subtitle=subtitle,
+ compounded=compounded,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def rolling_beta(returns, benchmark,
- window1=126, window1_label="6-Months",
- window2=252, window2_label="12-Months",
- lw=1.5, fontname='Arial', grayscale=False,
- figsize=(10, 3), ylabel=True,
- subtitle=True, savefig=None, show=True,
- prepare_returns=True):
+def rolling_beta(
+ returns,
+ benchmark,
+ window1=126,
+ window1_label="6-Months",
+ window2=252,
+ window2_label="12-Months",
+ lw=1.5,
+ fontname="Arial",
+ grayscale=False,
+ figsize=(10, 3),
+ ylabel=True,
+ subtitle=True,
+ savefig=None,
+ show=True,
+ prepare_returns=True,
+):
if prepare_returns:
returns = _utils._prepare_returns(returns)
benchmark = _utils._prepare_benchmark(benchmark, returns.index)
- fig = _core.plot_rolling_beta(returns, benchmark,
- window1=window1, window1_label=window1_label,
- window2=window2, window2_label=window2_label,
- title="Rolling Beta to Benchmark",
- fontname=fontname,
- grayscale=grayscale,
- lw=lw,
- figsize=figsize,
- ylabel=ylabel,
- subtitle=subtitle,
- savefig=savefig, show=show)
+ fig = _core.plot_rolling_beta(
+ returns,
+ benchmark,
+ window1=window1,
+ window1_label=window1_label,
+ window2=window2,
+ window2_label=window2_label,
+ title="Rolling Beta to Benchmark",
+ fontname=fontname,
+ grayscale=grayscale,
+ lw=lw,
+ figsize=figsize,
+ ylabel=ylabel,
+ subtitle=subtitle,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def rolling_volatility(returns, benchmark=None,
- period=126, period_label="6-Months",
- periods_per_year=252,
- lw=1.5, fontname='Arial', grayscale=False,
- figsize=(10, 3), ylabel="Volatility",
- subtitle=True, savefig=None, show=True):
+def rolling_volatility(
+ returns,
+ benchmark=None,
+ period=126,
+ period_label="6-Months",
+ periods_per_year=252,
+ lw=1.5,
+ fontname="Arial",
+ grayscale=False,
+ figsize=(10, 3),
+ ylabel="Volatility",
+ subtitle=True,
+ savefig=None,
+ show=True,
+):
returns = _stats.rolling_volatility(returns, period, periods_per_year)
if benchmark is not None:
benchmark = _utils._prepare_benchmark(benchmark, returns.index)
benchmark = _stats.rolling_volatility(
- benchmark, period, periods_per_year, prepare_returns=False)
-
- fig = _core.plot_rolling_stats(returns, benchmark,
- hline=returns.mean(),
- hlw=1.5,
- ylabel=ylabel,
- title='Rolling Volatility (%s)' % period_label,
- fontname=fontname,
- grayscale=grayscale,
- lw=lw,
- figsize=figsize,
- subtitle=subtitle,
- savefig=savefig, show=show)
+ benchmark, period, periods_per_year, prepare_returns=False
+ )
+
+ fig = _core.plot_rolling_stats(
+ returns,
+ benchmark,
+ hline=returns.mean(),
+ hlw=1.5,
+ ylabel=ylabel,
+ title="Rolling Volatility (%s)" % period_label,
+ fontname=fontname,
+ grayscale=grayscale,
+ lw=lw,
+ figsize=figsize,
+ subtitle=subtitle,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def rolling_sharpe(returns, benchmark=None, rf=0.,
- period=126, period_label="6-Months",
- periods_per_year=252,
- lw=1.25, fontname='Arial', grayscale=False,
- figsize=(10, 3), ylabel="Sharpe",
- subtitle=True, savefig=None, show=True):
+def rolling_sharpe(
+ returns,
+ benchmark=None,
+ rf=0.0,
+ period=126,
+ period_label="6-Months",
+ periods_per_year=252,
+ lw=1.25,
+ fontname="Arial",
+ grayscale=False,
+ figsize=(10, 3),
+ ylabel="Sharpe",
+ subtitle=True,
+ savefig=None,
+ show=True,
+):
returns = _stats.rolling_sharpe(
- returns, rf, period, True, periods_per_year, )
+ returns,
+ rf,
+ period,
+ True,
+ periods_per_year,
+ )
if benchmark is not None:
benchmark = _utils._prepare_benchmark(benchmark, returns.index, rf)
benchmark = _stats.rolling_sharpe(
- benchmark, rf, period, True, periods_per_year,
- prepare_returns=False)
-
- fig = _core.plot_rolling_stats(returns, benchmark,
- hline=returns.mean(),
- hlw=1.5,
- ylabel=ylabel,
- title='Rolling Sharpe (%s)' % period_label,
- fontname=fontname,
- grayscale=grayscale,
- lw=lw,
- figsize=figsize,
- subtitle=subtitle,
- savefig=savefig, show=show)
+ benchmark, rf, period, True, periods_per_year, prepare_returns=False
+ )
+
+ fig = _core.plot_rolling_stats(
+ returns,
+ benchmark,
+ hline=returns.mean(),
+ hlw=1.5,
+ ylabel=ylabel,
+ title="Rolling Sharpe (%s)" % period_label,
+ fontname=fontname,
+ grayscale=grayscale,
+ lw=lw,
+ figsize=figsize,
+ subtitle=subtitle,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def rolling_sortino(returns, benchmark=None, rf=0.,
- period=126, period_label="6-Months",
- periods_per_year=252,
- lw=1.25, fontname='Arial', grayscale=False,
- figsize=(10, 3), ylabel="Sortino",
- subtitle=True, savefig=None, show=True):
-
- returns = _stats.rolling_sortino(
- returns, rf, period, True, periods_per_year)
+def rolling_sortino(
+ returns,
+ benchmark=None,
+ rf=0.0,
+ period=126,
+ period_label="6-Months",
+ periods_per_year=252,
+ lw=1.25,
+ fontname="Arial",
+ grayscale=False,
+ figsize=(10, 3),
+ ylabel="Sortino",
+ subtitle=True,
+ savefig=None,
+ show=True,
+):
+
+ returns = _stats.rolling_sortino(returns, rf, period, True, periods_per_year)
if benchmark is not None:
benchmark = _utils._prepare_benchmark(benchmark, returns.index, rf)
benchmark = _stats.rolling_sortino(
- benchmark, rf, period, True, periods_per_year,
- prepare_returns=False)
-
- fig = _core.plot_rolling_stats(returns, benchmark,
- hline=returns.mean(),
- hlw=1.5,
- ylabel=ylabel,
- title='Rolling Sortino (%s)' % period_label,
- fontname=fontname,
- grayscale=grayscale,
- lw=lw,
- figsize=figsize,
- subtitle=subtitle,
- savefig=savefig, show=show)
+ benchmark, rf, period, True, periods_per_year, prepare_returns=False
+ )
+
+ fig = _core.plot_rolling_stats(
+ returns,
+ benchmark,
+ hline=returns.mean(),
+ hlw=1.5,
+ ylabel=ylabel,
+ title="Rolling Sortino (%s)" % period_label,
+ fontname=fontname,
+ grayscale=grayscale,
+ lw=lw,
+ figsize=figsize,
+ subtitle=subtitle,
+ savefig=savefig,
+ show=show,
+ )
if not show:
return fig
-def monthly_heatmap(returns, benchmark, annot_size=10, figsize=(10, 5),
- cbar=True, square=False, returns_label="Strategy",
- compounded=True, eoy=False,
- grayscale=False, fontname='Arial',
- ylabel=True, savefig=None, show=True, active=False):
+def monthly_heatmap(
+ returns,
+ benchmark,
+ annot_size=10,
+ figsize=(10, 5),
+ cbar=True,
+ square=False,
+ returns_label="Strategy",
+ compounded=True,
+ eoy=False,
+ grayscale=False,
+ fontname="Arial",
+ ylabel=True,
+ savefig=None,
+ show=True,
+ active=False,
+):
# colors, ls, alpha = _core._get_colors(grayscale)
- cmap = 'gray' if grayscale else 'RdYlGn'
+ cmap = "gray" if grayscale else "RdYlGn"
- returns = _stats.monthly_returns(returns, eoy=eoy,
- compounded=compounded) * 100
+ returns = _stats.monthly_returns(returns, eoy=eoy, compounded=compounded) * 100
fig_height = len(returns) / 2.5
@@ -696,49 +963,77 @@ def monthly_heatmap(returns, benchmark, annot_size=10, figsize=(10, 5),
figsize = (figsize[0], max([fig_height, figsize[1]]))
if cbar:
- figsize = (figsize[0]*1.051, max([fig_height, figsize[1]]))
+ figsize = (figsize[0] * 1.051, max([fig_height, figsize[1]]))
fig, ax = _plt.subplots(figsize=figsize)
- ax.spines['top'].set_visible(False)
- ax.spines['right'].set_visible(False)
- ax.spines['bottom'].set_visible(False)
- ax.spines['left'].set_visible(False)
+ ax.spines["top"].set_visible(False)
+ ax.spines["right"].set_visible(False)
+ ax.spines["bottom"].set_visible(False)
+ ax.spines["left"].set_visible(False)
- fig.set_facecolor('white')
- ax.set_facecolor('white')
+ fig.set_facecolor("white")
+ ax.set_facecolor("white")
# _sns.set(font_scale=.9)
if active and benchmark is not None:
- ax.set_title(f'{returns_label} - Monthly Active Returns (%)\n', fontsize=14, y=.995,
- fontname=fontname, fontweight='bold', color='black')
- benchmark = _stats.monthly_returns(benchmark, eoy=eoy,
- compounded=compounded) * 100
+ ax.set_title(
+ f"{returns_label} - Monthly Active Returns (%)\n",
+ fontsize=14,
+ y=0.995,
+ fontname=fontname,
+ fontweight="bold",
+ color="black",
+ )
+ benchmark = (
+ _stats.monthly_returns(benchmark, eoy=eoy, compounded=compounded) * 100
+ )
active_returns = returns - benchmark
- ax = _sns.heatmap(active_returns, ax=ax, annot=True, center=0,
- annot_kws={"size": annot_size},
- fmt="0.2f", linewidths=0.5,
- square=square, cbar=cbar, cmap=cmap,
- cbar_kws={'format': '%.0f%%'})
+ ax = _sns.heatmap(
+ active_returns,
+ ax=ax,
+ annot=True,
+ center=0,
+ annot_kws={"size": annot_size},
+ fmt="0.2f",
+ linewidths=0.5,
+ square=square,
+ cbar=cbar,
+ cmap=cmap,
+ cbar_kws={"format": "%.0f%%"},
+ )
else:
- ax.set_title(f'{returns_label} - Monthly Returns (%)\n', fontsize=14, y=.995,
- fontname=fontname, fontweight='bold', color='black')
- ax = _sns.heatmap(returns, ax=ax, annot=True, center=0,
- annot_kws={"size": annot_size},
- fmt="0.2f", linewidths=0.5,
- square=square, cbar=cbar, cmap=cmap,
- cbar_kws={'format': '%.0f%%'})
+ ax.set_title(
+ f"{returns_label} - Monthly Returns (%)\n",
+ fontsize=14,
+ y=0.995,
+ fontname=fontname,
+ fontweight="bold",
+ color="black",
+ )
+ ax = _sns.heatmap(
+ returns,
+ ax=ax,
+ annot=True,
+ center=0,
+ annot_kws={"size": annot_size},
+ fmt="0.2f",
+ linewidths=0.5,
+ square=square,
+ cbar=cbar,
+ cmap=cmap,
+ cbar_kws={"format": "%.0f%%"},
+ )
# _sns.set(font_scale=1)
# align plot to match other
if ylabel:
- ax.set_ylabel('Years', fontname=fontname,
- fontweight='bold', fontsize=12)
- ax.yaxis.set_label_coords(-.1, .5)
+ ax.set_ylabel("Years", fontname=fontname, fontweight="bold", fontsize=12)
+ ax.yaxis.set_label_coords(-0.1, 0.5)
ax.tick_params(colors="#808080")
- _plt.xticks(rotation=0, fontsize=annot_size*1.2)
- _plt.yticks(rotation=0, fontsize=annot_size*1.2)
+ _plt.xticks(rotation=0, fontsize=annot_size * 1.2)
+ _plt.yticks(rotation=0, fontsize=annot_size * 1.2)
try:
_plt.subplots_adjust(hspace=0, bottom=0, top=1)
@@ -766,13 +1061,31 @@ def monthly_heatmap(returns, benchmark, annot_size=10, figsize=(10, 5),
return None
-def monthly_returns(returns, annot_size=10, figsize=(10, 5),
- cbar=True, square=False,
- compounded=True, eoy=False,
- grayscale=False, fontname='Arial',
- ylabel=True, savefig=None, show=True):
- return monthly_heatmap(returns, annot_size, figsize,
- cbar, square,
- compounded, eoy,
- grayscale, fontname,
- ylabel, savefig, show)
+def monthly_returns(
+ returns,
+ annot_size=10,
+ figsize=(10, 5),
+ cbar=True,
+ square=False,
+ compounded=True,
+ eoy=False,
+ grayscale=False,
+ fontname="Arial",
+ ylabel=True,
+ savefig=None,
+ show=True,
+):
+ return monthly_heatmap(
+ returns,
+ annot_size,
+ figsize,
+ cbar,
+ square,
+ compounded,
+ eoy,
+ grayscale,
+ fontname,
+ ylabel,
+ savefig,
+ show,
+ )
diff --git a/quantstats/plots.py b/quantstats/plots.py
index d75ed5df..1ebe38ea 100644
--- a/quantstats/plots.py
+++ b/quantstats/plots.py
@@ -20,6 +20,7 @@
try:
from pandas.plotting import register_matplotlib_converters as _rmc
+
_rmc()
except ImportError:
pass
diff --git a/quantstats/reports.py b/quantstats/reports.py
index 8d529f05..5b983db3 100644
--- a/quantstats/reports.py
+++ b/quantstats/reports.py
@@ -21,30 +21,22 @@
import pandas as _pd
import numpy as _np
from math import sqrt as _sqrt, ceil as _ceil
-from datetime import (
- datetime as _dt, timedelta as _td
-)
+from datetime import datetime as _dt
from base64 import b64encode as _b64encode
import re as _regex
from tabulate import tabulate as _tabulate
-from . import (
- __version__, stats as _stats,
- utils as _utils, plots as _plots
-)
+from . import __version__, stats as _stats, utils as _utils, plots as _plots
from dateutil.relativedelta import relativedelta
from io import StringIO
+
try:
- from IPython.display import (
- display as iDisplay, HTML as iHTML
- )
+ from IPython.display import display as iDisplay, HTML as iHTML
except ImportError:
- from IPython.core.display import (
- display as iDisplay, HTML as iHTML
- )
+ from IPython.core.display import display as iDisplay, HTML as iHTML
def _get_trading_periods(periods_per_year=252):
- half_year = _ceil(periods_per_year/2)
+ half_year = _ceil(periods_per_year / 2)
return periods_per_year, half_year
@@ -59,10 +51,21 @@ def _match_dates(returns, benchmark):
return returns, benchmark
-def html(returns, benchmark=None, rf=0., grayscale=False,
- title='Strategy Tearsheet', output=None, compounded=True,
- periods_per_year=252, download_filename='quantstats-tearsheet.html',
- figfmt='svg', template_path=None, match_dates=True, **kwargs):
+def html(
+ returns,
+ benchmark=None,
+ rf=0.0,
+ grayscale=False,
+ title="Strategy Tearsheet",
+ output=None,
+ compounded=True,
+ periods_per_year=252,
+ download_filename="quantstats-tearsheet.html",
+ figfmt="svg",
+ template_path=None,
+ match_dates=True,
+ **kwargs,
+):
if output is None and not _utils._in_notebook():
raise ValueError("`output` must be specified")
@@ -73,7 +76,7 @@ def html(returns, benchmark=None, rf=0., grayscale=False,
win_year, win_half_year = _get_trading_periods(periods_per_year)
tpl = ""
- with open(template_path or __file__[:-4] + '.html') as f:
+ with open(template_path or __file__[:-4] + ".html") as f:
tpl = f.read()
f.close()
@@ -82,14 +85,14 @@ def html(returns, benchmark=None, rf=0., grayscale=False,
returns = returns.dropna()
returns = _utils._prepare_returns(returns)
- strategy_title = kwargs.get('strategy_title', 'Strategy')
+ strategy_title = kwargs.get("strategy_title", "Strategy")
if isinstance(returns, _pd.DataFrame):
if len(returns.columns) > 1 and isinstance(strategy_title, str):
strategy_title = list(returns.columns)
if benchmark is not None:
- benchmark_title = kwargs.get('benchmark_title', 'Benchmark')
- if kwargs.get('benchmark_title') is None:
+ benchmark_title = kwargs.get("benchmark_title", "Benchmark")
+ if kwargs.get("benchmark_title") is None:
if isinstance(benchmark, str):
benchmark_title = benchmark
elif isinstance(benchmark, _pd.Series):
@@ -97,17 +100,19 @@ def html(returns, benchmark=None, rf=0., grayscale=False,
elif isinstance(benchmark, _pd.DataFrame):
benchmark_title = benchmark[benchmark.columns[0]].name
- tpl = tpl.replace('{{benchmark_title}}', f"Benchmark is {benchmark_title.upper()} | ")
+ tpl = tpl.replace(
+ "{{benchmark_title}}", f"Benchmark is {benchmark_title.upper()} | "
+ )
benchmark = _utils._prepare_benchmark(benchmark, returns.index, rf)
if match_dates is True:
returns, benchmark = _match_dates(returns, benchmark)
else:
benchmark_title = None
- date_range = returns.index.strftime('%e %b, %Y')
- tpl = tpl.replace('{{date_range}}', date_range[0] + ' - ' + date_range[-1])
- tpl = tpl.replace('{{title}}', title)
- tpl = tpl.replace('{{v}}', __version__)
+ date_range = returns.index.strftime("%e %b, %Y")
+ tpl = tpl.replace("{{date_range}}", date_range[0] + " - " + date_range[-1])
+ tpl = tpl.replace("{{title}}", title)
+ tpl = tpl.replace("{{v}}", __version__)
if benchmark is not None:
benchmark.name = benchmark_title
@@ -116,245 +121,386 @@ def html(returns, benchmark=None, rf=0., grayscale=False,
elif isinstance(returns, _pd.DataFrame):
returns.columns = strategy_title
- mtrx = metrics(returns=returns, benchmark=benchmark,
- rf=rf, display=False, mode='full',
- sep=True, internal="True",
- compounded=compounded,
- periods_per_year=periods_per_year,
- prepare_returns=False,
- benchmark_title=benchmark_title,
- strategy_title=strategy_title)[2:]
-
- mtrx.index.name = 'Metric'
- tpl = tpl.replace('{{metrics}}', _html_table(mtrx))
+ mtrx = metrics(
+ returns=returns,
+ benchmark=benchmark,
+ rf=rf,
+ display=False,
+ mode="full",
+ sep=True,
+ internal="True",
+ compounded=compounded,
+ periods_per_year=periods_per_year,
+ prepare_returns=False,
+ benchmark_title=benchmark_title,
+ strategy_title=strategy_title,
+ )[2:]
+
+ mtrx.index.name = "Metric"
+ tpl = tpl.replace("{{metrics}}", _html_table(mtrx))
if isinstance(returns, _pd.DataFrame):
num_cols = len(returns.columns)
- for i in reversed(range(num_cols+1, num_cols + 3)):
+ for i in reversed(range(num_cols + 1, num_cols + 3)):
str_td = " | " * i
- tpl = tpl.replace(f'{str_td}
',
- '
|
'.format(i))
+ tpl = tpl.replace(
+ f"{str_td}
", '
|
'.format(i)
+ )
- tpl = tpl.replace(' | | |
',
- '
|
')
- tpl = tpl.replace(' | |
',
- '
|
')
+ tpl = tpl.replace(
+ " | | |
", '
|
'
+ )
+ tpl = tpl.replace(
+ " | |
", '
|
'
+ )
if benchmark is not None:
yoy = _stats.compare(
- returns, benchmark, "A", compounded=compounded,
- prepare_returns=False)
+ returns, benchmark, "A", compounded=compounded, prepare_returns=False
+ )
if isinstance(returns, _pd.Series):
- yoy.columns = [benchmark_title, strategy_title, 'Multiplier', 'Won']
+ yoy.columns = [benchmark_title, strategy_title, "Multiplier", "Won"]
elif isinstance(returns, _pd.DataFrame):
- yoy.columns = list(_pd.core.common.flatten([benchmark_title, strategy_title]))
- yoy.index.name = 'Year'
- tpl = tpl.replace('{{eoy_title}}', 'EOY Returns vs Benchmark
')
- tpl = tpl.replace('{{eoy_table}}', _html_table(yoy))
+ yoy.columns = list(
+ _pd.core.common.flatten([benchmark_title, strategy_title])
+ )
+ yoy.index.name = "Year"
+ tpl = tpl.replace("{{eoy_title}}", "EOY Returns vs Benchmark
")
+ tpl = tpl.replace("{{eoy_table}}", _html_table(yoy))
else:
# pct multiplier
- yoy = _pd.DataFrame(
- _utils.group_returns(returns, returns.index.year) * 100)
+ yoy = _pd.DataFrame(_utils.group_returns(returns, returns.index.year) * 100)
if isinstance(returns, _pd.Series):
- yoy.columns = ['Return']
- yoy['Cumulative'] = _utils.group_returns(
- returns, returns.index.year, True)
- yoy['Return'] = yoy['Return'].round(2).astype(str) + '%'
- yoy['Cumulative'] = (yoy['Cumulative'] *
- 100).round(2).astype(str) + '%'
+ yoy.columns = ["Return"]
+ yoy["Cumulative"] = _utils.group_returns(returns, returns.index.year, True)
+ yoy["Return"] = yoy["Return"].round(2).astype(str) + "%"
+ yoy["Cumulative"] = (yoy["Cumulative"] * 100).round(2).astype(str) + "%"
elif isinstance(returns, _pd.DataFrame):
# Don't show cumulative for multiple strategy portfolios
# just show compounded like when we have a benchmark
yoy.columns = list(_pd.core.common.flatten(strategy_title))
- yoy.index.name = 'Year'
- tpl = tpl.replace('{{eoy_title}}', 'EOY Returns
')
- tpl = tpl.replace('{{eoy_table}}', _html_table(yoy))
+ yoy.index.name = "Year"
+ tpl = tpl.replace("{{eoy_title}}", "EOY Returns
")
+ tpl = tpl.replace("{{eoy_table}}", _html_table(yoy))
if isinstance(returns, _pd.Series):
dd = _stats.to_drawdown_series(returns)
dd_info = _stats.drawdown_details(dd).sort_values(
- by='max drawdown', ascending=True)[:10]
- dd_info = dd_info[['start', 'end', 'max drawdown', 'days']]
- dd_info.columns = ['Started', 'Recovered', 'Drawdown', 'Days']
- tpl = tpl.replace('{{dd_info}}', _html_table(dd_info, False))
+ by="max drawdown", ascending=True
+ )[:10]
+ dd_info = dd_info[["start", "end", "max drawdown", "days"]]
+ dd_info.columns = ["Started", "Recovered", "Drawdown", "Days"]
+ tpl = tpl.replace("{{dd_info}}", _html_table(dd_info, False))
elif isinstance(returns, _pd.DataFrame):
dd_info_list = []
for col in returns.columns:
dd = _stats.to_drawdown_series(returns[col])
dd_info = _stats.drawdown_details(dd).sort_values(
- by='max drawdown', ascending=True)[:10]
- dd_info = dd_info[['start', 'end', 'max drawdown', 'days']]
- dd_info.columns = ['Started', 'Recovered', 'Drawdown', 'Days']
+ by="max drawdown", ascending=True
+ )[:10]
+ dd_info = dd_info[["start", "end", "max drawdown", "days"]]
+ dd_info.columns = ["Started", "Recovered", "Drawdown", "Days"]
dd_info_list.append(_html_table(dd_info, False))
dd_html_table = ""
for html_str, col in zip(dd_info_list, returns.columns):
- dd_html_table = dd_html_table + f"{col}
" + StringIO(html_str).read()
- tpl = tpl.replace('{{dd_info}}', dd_html_table)
+ dd_html_table = (
+ dd_html_table + f"{col}
" + StringIO(html_str).read()
+ )
+ tpl = tpl.replace("{{dd_info}}", dd_html_table)
- active = kwargs.get('active_returns', 'False')
+ active = kwargs.get("active_returns", "False")
# plots
figfile = _utils._file_stream()
- _plots.returns(returns, benchmark, grayscale=grayscale,
- figsize=(8, 5), subtitle=False,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, cumulative=compounded,
- prepare_returns=False)
- tpl = tpl.replace('{{returns}}', _embed_figure(figfile, figfmt))
+ _plots.returns(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(8, 5),
+ subtitle=False,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ cumulative=compounded,
+ prepare_returns=False,
+ )
+ tpl = tpl.replace("{{returns}}", _embed_figure(figfile, figfmt))
figfile = _utils._file_stream()
- _plots.log_returns(returns, benchmark, grayscale=grayscale,
- figsize=(8, 4), subtitle=False,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, cumulative=compounded,
- prepare_returns=False)
- tpl = tpl.replace('{{log_returns}}', _embed_figure(figfile, figfmt))
+ _plots.log_returns(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(8, 4),
+ subtitle=False,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ cumulative=compounded,
+ prepare_returns=False,
+ )
+ tpl = tpl.replace("{{log_returns}}", _embed_figure(figfile, figfmt))
if benchmark is not None:
figfile = _utils._file_stream()
- _plots.returns(returns, benchmark, match_volatility=True,
- grayscale=grayscale, figsize=(8, 4), subtitle=False,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, cumulative=compounded,
- prepare_returns=False)
- tpl = tpl.replace('{{vol_returns}}', _embed_figure(figfile, figfmt))
+ _plots.returns(
+ returns,
+ benchmark,
+ match_volatility=True,
+ grayscale=grayscale,
+ figsize=(8, 4),
+ subtitle=False,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ cumulative=compounded,
+ prepare_returns=False,
+ )
+ tpl = tpl.replace("{{vol_returns}}", _embed_figure(figfile, figfmt))
figfile = _utils._file_stream()
- _plots.yearly_returns(returns, benchmark, grayscale=grayscale,
- figsize=(8, 4), subtitle=False,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, compounded=compounded,
- prepare_returns=False)
- tpl = tpl.replace('{{eoy_returns}}', _embed_figure(figfile, figfmt))
+ _plots.yearly_returns(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(8, 4),
+ subtitle=False,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ compounded=compounded,
+ prepare_returns=False,
+ )
+ tpl = tpl.replace("{{eoy_returns}}", _embed_figure(figfile, figfmt))
figfile = _utils._file_stream()
- _plots.histogram(returns, benchmark, grayscale=grayscale,
- figsize=(7, 4), subtitle=False,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, compounded=compounded,
- prepare_returns=False)
- tpl = tpl.replace('{{monthly_dist}}', _embed_figure(figfile, figfmt))
+ _plots.histogram(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(7, 4),
+ subtitle=False,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ compounded=compounded,
+ prepare_returns=False,
+ )
+ tpl = tpl.replace("{{monthly_dist}}", _embed_figure(figfile, figfmt))
figfile = _utils._file_stream()
- _plots.daily_returns(returns, benchmark, grayscale=grayscale,
- figsize=(8, 3), subtitle=False,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False,
- prepare_returns=False, active=active)
- tpl = tpl.replace('{{daily_returns}}', _embed_figure(figfile, figfmt))
+ _plots.daily_returns(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(8, 3),
+ subtitle=False,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ prepare_returns=False,
+ active=active,
+ )
+ tpl = tpl.replace("{{daily_returns}}", _embed_figure(figfile, figfmt))
if benchmark is not None:
figfile = _utils._file_stream()
- _plots.rolling_beta(returns, benchmark, grayscale=grayscale,
- figsize=(8, 3), subtitle=False,
- window1=win_half_year, window2=win_year,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False,
- prepare_returns=False)
- tpl = tpl.replace('{{rolling_beta}}', _embed_figure(figfile, figfmt))
+ _plots.rolling_beta(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(8, 3),
+ subtitle=False,
+ window1=win_half_year,
+ window2=win_year,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ prepare_returns=False,
+ )
+ tpl = tpl.replace("{{rolling_beta}}", _embed_figure(figfile, figfmt))
figfile = _utils._file_stream()
- _plots.rolling_volatility(returns, benchmark, grayscale=grayscale,
- figsize=(8, 3), subtitle=False,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, period=win_half_year,
- periods_per_year=win_year)
- tpl = tpl.replace('{{rolling_vol}}', _embed_figure(figfile, figfmt))
+ _plots.rolling_volatility(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(8, 3),
+ subtitle=False,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ period=win_half_year,
+ periods_per_year=win_year,
+ )
+ tpl = tpl.replace("{{rolling_vol}}", _embed_figure(figfile, figfmt))
figfile = _utils._file_stream()
- _plots.rolling_sharpe(returns, grayscale=grayscale,
- figsize=(8, 3), subtitle=False,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, period=win_half_year,
- periods_per_year=win_year)
- tpl = tpl.replace('{{rolling_sharpe}}', _embed_figure(figfile, figfmt))
+ _plots.rolling_sharpe(
+ returns,
+ grayscale=grayscale,
+ figsize=(8, 3),
+ subtitle=False,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ period=win_half_year,
+ periods_per_year=win_year,
+ )
+ tpl = tpl.replace("{{rolling_sharpe}}", _embed_figure(figfile, figfmt))
figfile = _utils._file_stream()
- _plots.rolling_sortino(returns, grayscale=grayscale,
- figsize=(8, 3), subtitle=False,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, period=win_half_year,
- periods_per_year=win_year)
- tpl = tpl.replace('{{rolling_sortino}}', _embed_figure(figfile, figfmt))
+ _plots.rolling_sortino(
+ returns,
+ grayscale=grayscale,
+ figsize=(8, 3),
+ subtitle=False,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ period=win_half_year,
+ periods_per_year=win_year,
+ )
+ tpl = tpl.replace("{{rolling_sortino}}", _embed_figure(figfile, figfmt))
figfile = _utils._file_stream()
if isinstance(returns, _pd.Series):
- _plots.drawdowns_periods(returns, grayscale=grayscale,
- figsize=(8, 4), subtitle=False, title=returns.name,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, compounded=compounded,
- prepare_returns=False)
- tpl = tpl.replace('{{dd_periods}}', _embed_figure(figfile, figfmt))
+ _plots.drawdowns_periods(
+ returns,
+ grayscale=grayscale,
+ figsize=(8, 4),
+ subtitle=False,
+ title=returns.name,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ compounded=compounded,
+ prepare_returns=False,
+ )
+ tpl = tpl.replace("{{dd_periods}}", _embed_figure(figfile, figfmt))
elif isinstance(returns, _pd.DataFrame):
embed = []
for col in returns.columns:
- _plots.drawdowns_periods(returns[col], grayscale=grayscale,
- figsize=(8, 4), subtitle=False, title=col,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, compounded=compounded,
- prepare_returns=False)
+ _plots.drawdowns_periods(
+ returns[col],
+ grayscale=grayscale,
+ figsize=(8, 4),
+ subtitle=False,
+ title=col,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ compounded=compounded,
+ prepare_returns=False,
+ )
embed.append(figfile)
- tpl = tpl.replace('{{dd_periods}}', _embed_figure(embed, figfmt))
+ tpl = tpl.replace("{{dd_periods}}", _embed_figure(embed, figfmt))
figfile = _utils._file_stream()
- _plots.drawdown(returns, grayscale=grayscale,
- figsize=(8, 3), subtitle=False,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False)
- tpl = tpl.replace('{{dd_plot}}', _embed_figure(figfile, figfmt))
+ _plots.drawdown(
+ returns,
+ grayscale=grayscale,
+ figsize=(8, 3),
+ subtitle=False,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ )
+ tpl = tpl.replace("{{dd_plot}}", _embed_figure(figfile, figfmt))
figfile = _utils._file_stream()
if isinstance(returns, _pd.Series):
- _plots.monthly_heatmap(returns, benchmark, grayscale=grayscale,
- figsize=(8, 4), cbar=False, returns_label=returns.name,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, compounded=compounded, active=active)
- tpl = tpl.replace('{{monthly_heatmap}}', _embed_figure(figfile, figfmt))
+ _plots.monthly_heatmap(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(8, 4),
+ cbar=False,
+ returns_label=returns.name,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ compounded=compounded,
+ active=active,
+ )
+ tpl = tpl.replace("{{monthly_heatmap}}", _embed_figure(figfile, figfmt))
elif isinstance(returns, _pd.DataFrame):
embed = []
for col in returns.columns:
- _plots.monthly_heatmap(returns[col], benchmark, grayscale=grayscale,
- figsize=(8, 4), cbar=False, returns_label=col,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, compounded=compounded, active=active)
+ _plots.monthly_heatmap(
+ returns[col],
+ benchmark,
+ grayscale=grayscale,
+ figsize=(8, 4),
+ cbar=False,
+ returns_label=col,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ compounded=compounded,
+ active=active,
+ )
embed.append(figfile)
- tpl = tpl.replace('{{monthly_heatmap}}', _embed_figure(embed, figfmt))
+ tpl = tpl.replace("{{monthly_heatmap}}", _embed_figure(embed, figfmt))
figfile = _utils._file_stream()
if isinstance(returns, _pd.Series):
- _plots.distribution(returns, grayscale=grayscale,
- figsize=(8, 4), subtitle=False, title=returns.name,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, compounded=compounded,
- prepare_returns=False)
- tpl = tpl.replace('{{returns_dist}}', _embed_figure(figfile, figfmt))
+ _plots.distribution(
+ returns,
+ grayscale=grayscale,
+ figsize=(8, 4),
+ subtitle=False,
+ title=returns.name,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ compounded=compounded,
+ prepare_returns=False,
+ )
+ tpl = tpl.replace("{{returns_dist}}", _embed_figure(figfile, figfmt))
elif isinstance(returns, _pd.DataFrame):
embed = []
for col in returns.columns:
- _plots.distribution(returns[col], grayscale=grayscale,
- figsize=(8, 4), subtitle=False, title=col,
- savefig={'fname': figfile, 'format': figfmt},
- show=False, ylabel=False, compounded=compounded,
- prepare_returns=False)
+ _plots.distribution(
+ returns[col],
+ grayscale=grayscale,
+ figsize=(8, 4),
+ subtitle=False,
+ title=col,
+ savefig={"fname": figfile, "format": figfmt},
+ show=False,
+ ylabel=False,
+ compounded=compounded,
+ prepare_returns=False,
+ )
embed.append(figfile)
- tpl = tpl.replace('{{returns_dist}}', _embed_figure(embed, figfmt))
+ tpl = tpl.replace("{{returns_dist}}", _embed_figure(embed, figfmt))
- tpl = _regex.sub(r'\{\{(.*?)\}\}', '', tpl)
- tpl = tpl.replace('white-space:pre;', '')
+ tpl = _regex.sub(r"\{\{(.*?)\}\}", "", tpl)
+ tpl = tpl.replace("white-space:pre;", "")
if output is None:
# _open_html(tpl)
_download_html(tpl, download_filename)
return
- with open(output, 'w', encoding='utf-8') as f:
+ with open(output, "w", encoding="utf-8") as f:
f.write(tpl)
-def full(returns, benchmark=None, rf=0., grayscale=False,
- figsize=(8, 5), display=True, compounded=True,
- periods_per_year=252, match_dates=True, **kwargs):
+def full(
+ returns,
+ benchmark=None,
+ rf=0.0,
+ grayscale=False,
+ figsize=(8, 5),
+ display=True,
+ compounded=True,
+ periods_per_year=252,
+ match_dates=True,
+ **kwargs,
+):
# prepare timeseries
if match_dates:
@@ -367,9 +513,9 @@ def full(returns, benchmark=None, rf=0., grayscale=False,
benchmark_title = None
if benchmark is not None:
- benchmark_title = kwargs.get('benchmark_title', 'Benchmark')
- strategy_title = kwargs.get('strategy_title', 'Strategy')
- active = kwargs.get('active_returns', 'False')
+ benchmark_title = kwargs.get("benchmark_title", "Benchmark")
+ strategy_title = kwargs.get("strategy_title", "Strategy")
+ active = kwargs.get("active_returns", "False")
if isinstance(returns, _pd.DataFrame):
if len(returns.columns) > 1 and isinstance(strategy_title, str):
@@ -386,8 +532,7 @@ def full(returns, benchmark=None, rf=0., grayscale=False,
if isinstance(dd, _pd.Series):
col = _stats.drawdown_details(dd).columns[4]
- dd_info = _stats.drawdown_details(dd).sort_values(by=col,
- ascending=True)[:5]
+ dd_info = _stats.drawdown_details(dd).sort_values(by=col, ascending=True)[:5]
if not dd_info.empty:
dd_info.index = range(1, min(6, len(dd_info) + 1))
dd_info.columns = map(lambda x: str(x).title(), dd_info.columns)
@@ -395,22 +540,30 @@ def full(returns, benchmark=None, rf=0., grayscale=False,
col = _stats.drawdown_details(dd).columns.get_level_values(1)[4]
dd_info_dict = {}
for ptf in dd.columns:
- dd_info = _stats.drawdown_details(dd[ptf]).sort_values(by=col,
- ascending=True)[:5]
+ dd_info = _stats.drawdown_details(dd[ptf]).sort_values(
+ by=col, ascending=True
+ )[:5]
if not dd_info.empty:
dd_info.index = range(1, min(6, len(dd_info) + 1))
dd_info.columns = map(lambda x: str(x).title(), dd_info.columns)
dd_info_dict[ptf] = dd_info
if _utils._in_notebook():
- iDisplay(iHTML('Performance Metrics
'))
- iDisplay(metrics(returns=returns, benchmark=benchmark,
- rf=rf, display=display, mode='full',
- compounded=compounded,
- periods_per_year=periods_per_year,
- prepare_returns=False,
- benchmark_title=benchmark_title,
- strategy_title=strategy_title))
+ iDisplay(iHTML("Performance Metrics
"))
+ iDisplay(
+ metrics(
+ returns=returns,
+ benchmark=benchmark,
+ rf=rf,
+ display=display,
+ mode="full",
+ compounded=compounded,
+ periods_per_year=periods_per_year,
+ prepare_returns=False,
+ benchmark_title=benchmark_title,
+ strategy_title=strategy_title,
+ )
+ )
if isinstance(dd, _pd.Series):
iDisplay(iHTML('Worst 5 Drawdowns
'))
@@ -420,51 +573,84 @@ def full(returns, benchmark=None, rf=0., grayscale=False,
iDisplay(dd_info)
elif isinstance(dd, _pd.DataFrame):
for ptf, dd_info in dd_info_dict.items():
- iDisplay(iHTML('%s - Worst 5 Drawdowns
' % ptf))
+ iDisplay(
+ iHTML(
+ '%s - Worst 5 Drawdowns
'
+ % ptf
+ )
+ )
if dd_info.empty:
iDisplay(iHTML("(no drawdowns)
"))
else:
iDisplay(dd_info)
- iDisplay(iHTML('Strategy Visualization
'))
+ iDisplay(iHTML("Strategy Visualization
"))
else:
- print('[Performance Metrics]\n')
- metrics(returns=returns, benchmark=benchmark,
- rf=rf, display=display, mode='full',
- compounded=compounded,
- periods_per_year=periods_per_year,
- prepare_returns=False,
- benchmark_title=benchmark_title,
- strategy_title=strategy_title)
- print('\n\n')
- print('[Worst 5 Drawdowns]\n')
+ print("[Performance Metrics]\n")
+ metrics(
+ returns=returns,
+ benchmark=benchmark,
+ rf=rf,
+ display=display,
+ mode="full",
+ compounded=compounded,
+ periods_per_year=periods_per_year,
+ prepare_returns=False,
+ benchmark_title=benchmark_title,
+ strategy_title=strategy_title,
+ )
+ print("\n\n")
+ print("[Worst 5 Drawdowns]\n")
if isinstance(dd, _pd.Series):
if dd_info.empty:
print("(no drawdowns)")
else:
- print(_tabulate(dd_info, headers="keys",
- tablefmt='simple', floatfmt=".2f"))
+ print(
+ _tabulate(
+ dd_info, headers="keys", tablefmt="simple", floatfmt=".2f"
+ )
+ )
elif isinstance(dd, _pd.DataFrame):
for ptf, dd_info in dd_info_dict.items():
if dd_info.empty:
print("(no drawdowns)")
else:
print(f"{ptf}\n")
- print(_tabulate(dd_info, headers="keys",
- tablefmt='simple', floatfmt=".2f"))
-
- print('\n\n')
- print('[Strategy Visualization]\nvia Matplotlib')
-
- plots(returns=returns, benchmark=benchmark,
- grayscale=grayscale, figsize=figsize, mode='full',
- periods_per_year=periods_per_year, prepare_returns=False,
- benchmark_title=benchmark_title, strategy_title=strategy_title, active=active)
+ print(
+ _tabulate(
+ dd_info, headers="keys", tablefmt="simple", floatfmt=".2f"
+ )
+ )
+
+ print("\n\n")
+ print("[Strategy Visualization]\nvia Matplotlib")
+
+ plots(
+ returns=returns,
+ benchmark=benchmark,
+ grayscale=grayscale,
+ figsize=figsize,
+ mode="full",
+ periods_per_year=periods_per_year,
+ prepare_returns=False,
+ benchmark_title=benchmark_title,
+ strategy_title=strategy_title,
+ active=active,
+ )
-def basic(returns, benchmark=None, rf=0., grayscale=False,
- figsize=(8, 5), display=True, compounded=True,
- periods_per_year=252, match_dates=True, **kwargs):
+def basic(
+ returns,
+ benchmark=None,
+ rf=0.0,
+ grayscale=False,
+ figsize=(8, 5),
+ display=True,
+ compounded=True,
+ periods_per_year=252,
+ match_dates=True,
+ **kwargs,
+):
# prepare timeseries
if match_dates:
@@ -477,46 +663,74 @@ def basic(returns, benchmark=None, rf=0., grayscale=False,
benchmark_title = None
if benchmark is not None:
- benchmark_title = kwargs.get('benchmark_title', 'Benchmark')
- strategy_title = kwargs.get('strategy_title', 'Strategy')
- active = kwargs.get('active_returns', 'False')
+ benchmark_title = kwargs.get("benchmark_title", "Benchmark")
+ strategy_title = kwargs.get("strategy_title", "Strategy")
+ active = kwargs.get("active_returns", "False")
if isinstance(returns, _pd.DataFrame):
if len(returns.columns) > 1 and isinstance(strategy_title, str):
strategy_title = list(returns.columns)
if _utils._in_notebook():
- iDisplay(iHTML('Performance Metrics
'))
- metrics(returns=returns, benchmark=benchmark,
- rf=rf, display=display, mode='basic',
- compounded=compounded,
- periods_per_year=periods_per_year,
- prepare_returns=False,
- benchmark_title=benchmark_title, strategy_title=strategy_title)
- iDisplay(iHTML('Strategy Visualization
'))
+ iDisplay(iHTML("Performance Metrics
"))
+ metrics(
+ returns=returns,
+ benchmark=benchmark,
+ rf=rf,
+ display=display,
+ mode="basic",
+ compounded=compounded,
+ periods_per_year=periods_per_year,
+ prepare_returns=False,
+ benchmark_title=benchmark_title,
+ strategy_title=strategy_title,
+ )
+ iDisplay(iHTML("Strategy Visualization
"))
else:
- print('[Performance Metrics]\n')
- metrics(returns=returns, benchmark=benchmark,
- rf=rf, display=display, mode='basic',
- compounded=compounded,
- periods_per_year=periods_per_year,
- prepare_returns=False,
- benchmark_title=benchmark_title, strategy_title=strategy_title)
-
- print('\n\n')
- print('[Strategy Visualization]\nvia Matplotlib')
-
- plots(returns=returns, benchmark=benchmark,
- grayscale=grayscale, figsize=figsize, mode='basic',
- periods_per_year=periods_per_year,
- prepare_returns=False, benchmark_title=benchmark_title,
- strategy_title=strategy_title, active=active)
+ print("[Performance Metrics]\n")
+ metrics(
+ returns=returns,
+ benchmark=benchmark,
+ rf=rf,
+ display=display,
+ mode="basic",
+ compounded=compounded,
+ periods_per_year=periods_per_year,
+ prepare_returns=False,
+ benchmark_title=benchmark_title,
+ strategy_title=strategy_title,
+ )
+
+ print("\n\n")
+ print("[Strategy Visualization]\nvia Matplotlib")
+
+ plots(
+ returns=returns,
+ benchmark=benchmark,
+ grayscale=grayscale,
+ figsize=figsize,
+ mode="basic",
+ periods_per_year=periods_per_year,
+ prepare_returns=False,
+ benchmark_title=benchmark_title,
+ strategy_title=strategy_title,
+ active=active,
+ )
-def metrics(returns, benchmark=None, rf=0., display=True,
- mode='basic', sep=False, compounded=True,
- periods_per_year=252, prepare_returns=True,
- match_dates=True, **kwargs):
+def metrics(
+ returns,
+ benchmark=None,
+ rf=0.0,
+ display=True,
+ mode="basic",
+ sep=False,
+ compounded=True,
+ periods_per_year=252,
+ prepare_returns=True,
+ match_dates=True,
+ **kwargs,
+):
if match_dates:
returns = returns.dropna()
@@ -528,18 +742,20 @@ def metrics(returns, benchmark=None, rf=0., display=True,
if benchmark is not None:
if isinstance(benchmark, str):
- benchmark_colname = f'Benchmark ({benchmark.upper()})'
+ benchmark_colname = f"Benchmark ({benchmark.upper()})"
elif isinstance(benchmark, _pd.DataFrame) and len(benchmark.columns) > 1:
- raise ValueError("`benchmark` must be a pandas Series, "
- "but a multi-column DataFrame was passed")
+ raise ValueError(
+ "`benchmark` must be a pandas Series, "
+ "but a multi-column DataFrame was passed"
+ )
if isinstance(returns, _pd.DataFrame):
if len(returns.columns) > 1:
- blank = [''] * len(returns.columns)
+ blank = [""] * len(returns.columns)
if isinstance(strategy_colname, str):
strategy_colname = list(returns.columns)
else:
- blank = ['']
+ blank = [""]
# if isinstance(returns, _pd.DataFrame):
# if len(returns.columns) > 1:
@@ -552,8 +768,12 @@ def metrics(returns, benchmark=None, rf=0., display=True,
if isinstance(returns, _pd.Series):
df = _pd.DataFrame({"returns": returns})
elif isinstance(returns, _pd.DataFrame):
- df = _pd.DataFrame({"returns_" + str(i + 1): returns[strategy_col]
- for i, strategy_col in enumerate(returns.columns)})
+ df = _pd.DataFrame(
+ {
+ "returns_" + str(i + 1): returns[strategy_col]
+ for i, strategy_col in enumerate(returns.columns)
+ }
+ )
if benchmark is not None:
benchmark = _utils._prepare_benchmark(benchmark, returns.index, rf)
@@ -561,27 +781,33 @@ def metrics(returns, benchmark=None, rf=0., display=True,
returns, benchmark = _match_dates(returns, benchmark)
df["benchmark"] = benchmark
if isinstance(returns, _pd.Series):
- blank = ['', '']
+ blank = ["", ""]
df["returns"] = returns
elif isinstance(returns, _pd.DataFrame):
- blank = [''] * len(returns.columns) + ['']
+ blank = [""] * len(returns.columns) + [""]
for i, strategy_col in enumerate(returns.columns):
- df["returns_" + str(i+1)] = returns[strategy_col]
+ df["returns_" + str(i + 1)] = returns[strategy_col]
if isinstance(returns, _pd.Series):
- s_start = {'returns': df['returns'].index.strftime('%Y-%m-%d')[0]}
- s_end = {'returns': df['returns'].index.strftime('%Y-%m-%d')[-1]}
- s_rf = {'returns': rf}
+ s_start = {"returns": df["returns"].index.strftime("%Y-%m-%d")[0]}
+ s_end = {"returns": df["returns"].index.strftime("%Y-%m-%d")[-1]}
+ s_rf = {"returns": rf}
elif isinstance(returns, _pd.DataFrame):
df_strategy_columns = [col for col in df.columns if col != "benchmark"]
- s_start = {strategy_col: df[strategy_col].dropna().index.strftime('%Y-%m-%d')[0] for strategy_col in df_strategy_columns}
- s_end = {strategy_col: df[strategy_col].dropna().index.strftime('%Y-%m-%d')[-1] for strategy_col in df_strategy_columns}
+ s_start = {
+ strategy_col: df[strategy_col].dropna().index.strftime("%Y-%m-%d")[0]
+ for strategy_col in df_strategy_columns
+ }
+ s_end = {
+ strategy_col: df[strategy_col].dropna().index.strftime("%Y-%m-%d")[-1]
+ for strategy_col in df_strategy_columns
+ }
s_rf = {strategy_col: rf for strategy_col in df_strategy_columns}
if "benchmark" in df:
- s_start['benchmark'] = df['benchmark'].index.strftime('%Y-%m-%d')[0]
- s_end['benchmark'] = df['benchmark'].index.strftime('%Y-%m-%d')[-1]
- s_rf['benchmark'] = rf
+ s_start["benchmark"] = df["benchmark"].index.strftime("%Y-%m-%d")[0]
+ s_end["benchmark"] = df["benchmark"].index.strftime("%Y-%m-%d")[-1]
+ s_rf["benchmark"] = rf
df = df.fillna(0)
@@ -591,200 +817,295 @@ def metrics(returns, benchmark=None, rf=0., display=True,
pct = 100
# return df
- dd = _calc_dd(df, display=(display or "internal" in kwargs),
- as_pct=kwargs.get("as_pct", False))
+ dd = _calc_dd(
+ df,
+ display=(display or "internal" in kwargs),
+ as_pct=kwargs.get("as_pct", False),
+ )
metrics = _pd.DataFrame()
- metrics['Start Period'] = _pd.Series(s_start)
- metrics['End Period'] = _pd.Series(s_end)
- metrics['Risk-Free Rate %'] = _pd.Series(s_rf)*100
- metrics['Time in Market %'] = _stats.exposure(df, prepare_returns=False) * pct
+ metrics["Start Period"] = _pd.Series(s_start)
+ metrics["End Period"] = _pd.Series(s_end)
+ metrics["Risk-Free Rate %"] = _pd.Series(s_rf) * 100
+ metrics["Time in Market %"] = _stats.exposure(df, prepare_returns=False) * pct
- metrics['~'] = blank
+ metrics["~"] = blank
if compounded:
- metrics['Cumulative Return %'] = (_stats.comp(df) * pct).map('{:,.2f}'.format)
+ metrics["Cumulative Return %"] = (_stats.comp(df) * pct).map("{:,.2f}".format)
else:
- metrics['Total Return %'] = (df.sum() * pct).map('{:,.2f}'.format)
+ metrics["Total Return %"] = (df.sum() * pct).map("{:,.2f}".format)
- metrics['CAGR﹪%'] = _stats.cagr(df, rf, compounded) * pct
+ metrics["CAGR﹪%"] = _stats.cagr(df, rf, compounded) * pct
- metrics['~~~~~~~~~~~~~~'] = blank
+ metrics["~~~~~~~~~~~~~~"] = blank
- metrics['Sharpe'] = _stats.sharpe(df, rf, win_year, True)
- metrics['Prob. Sharpe Ratio %'] = _stats.probabilistic_sharpe_ratio(df, rf, win_year, False) * pct
- if mode.lower() == 'full':
- metrics['Smart Sharpe'] = _stats.smart_sharpe(df, rf, win_year, True)
+ metrics["Sharpe"] = _stats.sharpe(df, rf, win_year, True)
+ metrics["Prob. Sharpe Ratio %"] = (
+ _stats.probabilistic_sharpe_ratio(df, rf, win_year, False) * pct
+ )
+ if mode.lower() == "full":
+ metrics["Smart Sharpe"] = _stats.smart_sharpe(df, rf, win_year, True)
# metrics['Prob. Smart Sharpe Ratio %'] = _stats.probabilistic_sharpe_ratio(df, rf, win_year, False, True) * pct
- metrics['Sortino'] = _stats.sortino(df, rf, win_year, True)
- if mode.lower() == 'full':
+ metrics["Sortino"] = _stats.sortino(df, rf, win_year, True)
+ if mode.lower() == "full":
# metrics['Prob. Sortino Ratio %'] = _stats.probabilistic_sortino_ratio(df, rf, win_year, False) * pct
- metrics['Smart Sortino'] = _stats.smart_sortino(df, rf, win_year, True)
+ metrics["Smart Sortino"] = _stats.smart_sortino(df, rf, win_year, True)
# metrics['Prob. Smart Sortino Ratio %'] = _stats.probabilistic_sortino_ratio(df, rf, win_year, False, True) * pct
- metrics['Sortino/√2'] = metrics['Sortino'] / _sqrt(2)
- if mode.lower() == 'full':
+ metrics["Sortino/√2"] = metrics["Sortino"] / _sqrt(2)
+ if mode.lower() == "full":
# metrics['Prob. Sortino/√2 Ratio %'] = _stats.probabilistic_adjusted_sortino_ratio(df, rf, win_year, False) * pct
- metrics['Smart Sortino/√2'] = metrics['Smart Sortino'] / _sqrt(2)
+ metrics["Smart Sortino/√2"] = metrics["Smart Sortino"] / _sqrt(2)
# metrics['Prob. Smart Sortino/√2 Ratio %'] = _stats.probabilistic_adjusted_sortino_ratio(df, rf, win_year, False, True) * pct
- metrics['Omega'] = _stats.omega(df, rf, 0., win_year)
+ metrics["Omega"] = _stats.omega(df, rf, 0.0, win_year)
- metrics['~~~~~~~~'] = blank
- metrics['Max Drawdown %'] = blank
- metrics['Longest DD Days'] = blank
+ metrics["~~~~~~~~"] = blank
+ metrics["Max Drawdown %"] = blank
+ metrics["Longest DD Days"] = blank
- if mode.lower() == 'full':
+ if mode.lower() == "full":
if isinstance(returns, _pd.Series):
- ret_vol = _stats.volatility(
- df['returns'], win_year, True, prepare_returns=False) * pct
+ ret_vol = (
+ _stats.volatility(df["returns"], win_year, True, prepare_returns=False)
+ * pct
+ )
elif isinstance(returns, _pd.DataFrame):
- ret_vol = ([_stats.volatility(df[strategy_col], win_year, True, prepare_returns=False) * pct
- for strategy_col in df_strategy_columns])
+ ret_vol = [
+ _stats.volatility(
+ df[strategy_col], win_year, True, prepare_returns=False
+ )
+ * pct
+ for strategy_col in df_strategy_columns
+ ]
if "benchmark" in df:
- bench_vol = _stats.volatility(
- df['benchmark'], win_year, True, prepare_returns=False) * pct
+ bench_vol = (
+ _stats.volatility(
+ df["benchmark"], win_year, True, prepare_returns=False
+ )
+ * pct
+ )
vol_ = [ret_vol, bench_vol]
if isinstance(ret_vol, list):
- metrics['Volatility (ann.) %'] = list(_pd.core.common.flatten(vol_))
+ metrics["Volatility (ann.) %"] = list(_pd.core.common.flatten(vol_))
else:
- metrics['Volatility (ann.) %'] = vol_
+ metrics["Volatility (ann.) %"] = vol_
if isinstance(returns, _pd.Series):
- metrics['R^2'] = _stats.r_squared(
- df['returns'], df['benchmark'], prepare_returns=False)
- metrics['Information Ratio'] = _stats.information_ratio(
- df['returns'], df['benchmark'], prepare_returns=False)
+ metrics["R^2"] = _stats.r_squared(
+ df["returns"], df["benchmark"], prepare_returns=False
+ )
+ metrics["Information Ratio"] = _stats.information_ratio(
+ df["returns"], df["benchmark"], prepare_returns=False
+ )
elif isinstance(returns, _pd.DataFrame):
- metrics['R^2'] = ([_stats.r_squared(df[strategy_col], df['benchmark'], prepare_returns=False).round(2)
- for strategy_col in df_strategy_columns]) + ['-']
- metrics['Information Ratio'] = ([_stats.information_ratio(df[strategy_col], df['benchmark'], prepare_returns=False).round(2)
- for strategy_col in df_strategy_columns]) + ['-']
+ metrics["R^2"] = (
+ [
+ _stats.r_squared(
+ df[strategy_col], df["benchmark"], prepare_returns=False
+ ).round(2)
+ for strategy_col in df_strategy_columns
+ ]
+ ) + ["-"]
+ metrics["Information Ratio"] = (
+ [
+ _stats.information_ratio(
+ df[strategy_col], df["benchmark"], prepare_returns=False
+ ).round(2)
+ for strategy_col in df_strategy_columns
+ ]
+ ) + ["-"]
else:
if isinstance(returns, _pd.Series):
- metrics['Volatility (ann.) %'] = [ret_vol]
+ metrics["Volatility (ann.) %"] = [ret_vol]
elif isinstance(returns, _pd.DataFrame):
- metrics['Volatility (ann.) %'] = ret_vol
-
- metrics['Calmar'] = _stats.calmar(df, prepare_returns=False)
- metrics['Skew'] = _stats.skew(df, prepare_returns=False)
- metrics['Kurtosis'] = _stats.kurtosis(df, prepare_returns=False)
-
- metrics['~~~~~~~~~~'] = blank
-
- metrics['Expected Daily %%'] = _stats.expected_return(
- df, prepare_returns=False) * pct
- metrics['Expected Monthly %%'] = _stats.expected_return(
- df, aggregate='M', prepare_returns=False) * pct
- metrics['Expected Yearly %%'] = _stats.expected_return(
- df, aggregate='A', prepare_returns=False) * pct
- metrics['Kelly Criterion %'] = _stats.kelly_criterion(
- df, prepare_returns=False) * pct
- metrics['Risk of Ruin %'] = _stats.risk_of_ruin(
- df, prepare_returns=False)
-
- metrics['Daily Value-at-Risk %'] = -abs(_stats.var(
- df, prepare_returns=False) * pct)
- metrics['Expected Shortfall (cVaR) %'] = -abs(_stats.cvar(
- df, prepare_returns=False) * pct)
-
- metrics['~~~~~~'] = blank
-
- if mode.lower() == 'full':
- metrics['Max Consecutive Wins *int'] = _stats.consecutive_wins(df)
- metrics['Max Consecutive Losses *int'] = _stats.consecutive_losses(df)
-
- metrics['Gain/Pain Ratio'] = _stats.gain_to_pain_ratio(df, rf)
- metrics['Gain/Pain (1M)'] = _stats.gain_to_pain_ratio(df, rf, "M")
+ metrics["Volatility (ann.) %"] = ret_vol
+
+ metrics["Calmar"] = _stats.calmar(df, prepare_returns=False)
+ metrics["Skew"] = _stats.skew(df, prepare_returns=False)
+ metrics["Kurtosis"] = _stats.kurtosis(df, prepare_returns=False)
+
+ metrics["~~~~~~~~~~"] = blank
+
+ metrics["Expected Daily %%"] = (
+ _stats.expected_return(df, prepare_returns=False) * pct
+ )
+ metrics["Expected Monthly %%"] = (
+ _stats.expected_return(df, aggregate="M", prepare_returns=False) * pct
+ )
+ metrics["Expected Yearly %%"] = (
+ _stats.expected_return(df, aggregate="A", prepare_returns=False) * pct
+ )
+ metrics["Kelly Criterion %"] = (
+ _stats.kelly_criterion(df, prepare_returns=False) * pct
+ )
+ metrics["Risk of Ruin %"] = _stats.risk_of_ruin(df, prepare_returns=False)
+
+ metrics["Daily Value-at-Risk %"] = -abs(
+ _stats.var(df, prepare_returns=False) * pct
+ )
+ metrics["Expected Shortfall (cVaR) %"] = -abs(
+ _stats.cvar(df, prepare_returns=False) * pct
+ )
+
+ metrics["~~~~~~"] = blank
+
+ if mode.lower() == "full":
+ metrics["Max Consecutive Wins *int"] = _stats.consecutive_wins(df)
+ metrics["Max Consecutive Losses *int"] = _stats.consecutive_losses(df)
+
+ metrics["Gain/Pain Ratio"] = _stats.gain_to_pain_ratio(df, rf)
+ metrics["Gain/Pain (1M)"] = _stats.gain_to_pain_ratio(df, rf, "M")
# if mode.lower() == 'full':
# metrics['GPR (3M)'] = _stats.gain_to_pain_ratio(df, rf, "Q")
# metrics['GPR (6M)'] = _stats.gain_to_pain_ratio(df, rf, "2Q")
# metrics['GPR (1Y)'] = _stats.gain_to_pain_ratio(df, rf, "A")
- metrics['~~~~~~~'] = blank
+ metrics["~~~~~~~"] = blank
- metrics['Payoff Ratio'] = _stats.payoff_ratio(df, prepare_returns=False)
- metrics['Profit Factor'] = _stats.profit_factor(df, prepare_returns=False)
- metrics['Common Sense Ratio'] = _stats.common_sense_ratio(df, prepare_returns=False)
- metrics['CPC Index'] = _stats.cpc_index(df, prepare_returns=False)
- metrics['Tail Ratio'] = _stats.tail_ratio(df, prepare_returns=False)
- metrics['Outlier Win Ratio'] = _stats.outlier_win_ratio(df, prepare_returns=False)
- metrics['Outlier Loss Ratio'] = _stats.outlier_loss_ratio(df, prepare_returns=False)
+ metrics["Payoff Ratio"] = _stats.payoff_ratio(df, prepare_returns=False)
+ metrics["Profit Factor"] = _stats.profit_factor(df, prepare_returns=False)
+ metrics["Common Sense Ratio"] = _stats.common_sense_ratio(df, prepare_returns=False)
+ metrics["CPC Index"] = _stats.cpc_index(df, prepare_returns=False)
+ metrics["Tail Ratio"] = _stats.tail_ratio(df, prepare_returns=False)
+ metrics["Outlier Win Ratio"] = _stats.outlier_win_ratio(df, prepare_returns=False)
+ metrics["Outlier Loss Ratio"] = _stats.outlier_loss_ratio(df, prepare_returns=False)
# returns
- metrics['~~'] = blank
+ metrics["~~"] = blank
comp_func = _stats.comp if compounded else _np.sum
today = df.index[-1] # _dt.today()
- metrics['MTD %'] = comp_func(df[df.index >= _dt(today.year, today.month, 1)]) * pct
+ metrics["MTD %"] = comp_func(df[df.index >= _dt(today.year, today.month, 1)]) * pct
d = today - relativedelta(months=3)
- metrics['3M %'] = comp_func(df[df.index >= d]) * pct
+ metrics["3M %"] = comp_func(df[df.index >= d]) * pct
d = today - relativedelta(months=6)
- metrics['6M %'] = comp_func(df[df.index >= d]) * pct
+ metrics["6M %"] = comp_func(df[df.index >= d]) * pct
- metrics['YTD %'] = comp_func(df[df.index >= _dt(today.year, 1, 1)]) * pct
+ metrics["YTD %"] = comp_func(df[df.index >= _dt(today.year, 1, 1)]) * pct
d = today - relativedelta(years=1)
- metrics['1Y %'] = comp_func(df[df.index >= d]) * pct
+ metrics["1Y %"] = comp_func(df[df.index >= d]) * pct
d = today - relativedelta(months=35)
- metrics['3Y (ann.) %'] = _stats.cagr(df[df.index >= d], 0., compounded) * pct
+ metrics["3Y (ann.) %"] = _stats.cagr(df[df.index >= d], 0.0, compounded) * pct
d = today - relativedelta(months=59)
- metrics['5Y (ann.) %'] = _stats.cagr(df[df.index >= d], 0., compounded) * pct
+ metrics["5Y (ann.) %"] = _stats.cagr(df[df.index >= d], 0.0, compounded) * pct
d = today - relativedelta(years=10)
- metrics['10Y (ann.) %'] = _stats.cagr(df[df.index >= d], 0., compounded) * pct
+ metrics["10Y (ann.) %"] = _stats.cagr(df[df.index >= d], 0.0, compounded) * pct
- metrics['All-time (ann.) %'] = _stats.cagr(df, 0., compounded) * pct
+ metrics["All-time (ann.) %"] = _stats.cagr(df, 0.0, compounded) * pct
# best/worst
- if mode.lower() == 'full':
- metrics['~~~'] = blank
- metrics['Best Day %'] = _stats.best(df, prepare_returns=False) * pct
- metrics['Worst Day %'] = _stats.worst(df, prepare_returns=False) * pct
- metrics['Best Month %'] = _stats.best(df, aggregate='M', prepare_returns=False) * pct
- metrics['Worst Month %'] = _stats.worst(df, aggregate='M', prepare_returns=False) * pct
- metrics['Best Year %'] = _stats.best(df, aggregate='A', prepare_returns=False) * pct
- metrics['Worst Year %'] = _stats.worst(df, aggregate='A', prepare_returns=False) * pct
+ if mode.lower() == "full":
+ metrics["~~~"] = blank
+ metrics["Best Day %"] = _stats.best(df, prepare_returns=False) * pct
+ metrics["Worst Day %"] = _stats.worst(df, prepare_returns=False) * pct
+ metrics["Best Month %"] = (
+ _stats.best(df, aggregate="M", prepare_returns=False) * pct
+ )
+ metrics["Worst Month %"] = (
+ _stats.worst(df, aggregate="M", prepare_returns=False) * pct
+ )
+ metrics["Best Year %"] = (
+ _stats.best(df, aggregate="A", prepare_returns=False) * pct
+ )
+ metrics["Worst Year %"] = (
+ _stats.worst(df, aggregate="A", prepare_returns=False) * pct
+ )
# dd
- metrics['~~~~'] = blank
+ metrics["~~~~"] = blank
for ix, row in dd.iterrows():
metrics[ix] = row
- metrics['Recovery Factor'] = _stats.recovery_factor(df)
- metrics['Ulcer Index'] = _stats.ulcer_index(df)
- metrics['Serenity Index'] = _stats.serenity_index(df, rf)
+ metrics["Recovery Factor"] = _stats.recovery_factor(df)
+ metrics["Ulcer Index"] = _stats.ulcer_index(df)
+ metrics["Serenity Index"] = _stats.serenity_index(df, rf)
# win rate
- if mode.lower() == 'full':
- metrics['~~~~~'] = blank
- metrics['Avg. Up Month %'] = _stats.avg_win(df, aggregate='M', prepare_returns=False) * pct
- metrics['Avg. Down Month %'] = _stats.avg_loss(df, aggregate='M', prepare_returns=False) * pct
- metrics['Win Days %%'] = _stats.win_rate(df, prepare_returns=False) * pct
- metrics['Win Month %%'] = _stats.win_rate(df, aggregate='M', prepare_returns=False) * pct
- metrics['Win Quarter %%'] = _stats.win_rate(df, aggregate='Q', prepare_returns=False) * pct
- metrics['Win Year %%'] = _stats.win_rate(df, aggregate='A', prepare_returns=False) * pct
+ if mode.lower() == "full":
+ metrics["~~~~~"] = blank
+ metrics["Avg. Up Month %"] = (
+ _stats.avg_win(df, aggregate="M", prepare_returns=False) * pct
+ )
+ metrics["Avg. Down Month %"] = (
+ _stats.avg_loss(df, aggregate="M", prepare_returns=False) * pct
+ )
+ metrics["Win Days %%"] = _stats.win_rate(df, prepare_returns=False) * pct
+ metrics["Win Month %%"] = (
+ _stats.win_rate(df, aggregate="M", prepare_returns=False) * pct
+ )
+ metrics["Win Quarter %%"] = (
+ _stats.win_rate(df, aggregate="Q", prepare_returns=False) * pct
+ )
+ metrics["Win Year %%"] = (
+ _stats.win_rate(df, aggregate="A", prepare_returns=False) * pct
+ )
if "benchmark" in df:
- metrics['~~~~~~~~~~~~'] = blank
+ metrics["~~~~~~~~~~~~"] = blank
if isinstance(returns, _pd.Series):
- greeks = _stats.greeks(df['returns'], df['benchmark'], win_year, prepare_returns=False)
- metrics['Beta'] = [str(round(greeks['beta'], 2)), '-']
- metrics['Alpha'] = [str(round(greeks['alpha'], 2)), '-']
- metrics['Correlation'] = [str(round(df['benchmark'].corr(df['returns']) * pct, 2))+'%', '-']
- metrics['Treynor Ratio'] = [str(round(_stats.treynor_ratio(df['returns'], df['benchmark'], win_year, rf)*pct, 2))+'%', '-']
+ greeks = _stats.greeks(
+ df["returns"], df["benchmark"], win_year, prepare_returns=False
+ )
+ metrics["Beta"] = [str(round(greeks["beta"], 2)), "-"]
+ metrics["Alpha"] = [str(round(greeks["alpha"], 2)), "-"]
+ metrics["Correlation"] = [
+ str(round(df["benchmark"].corr(df["returns"]) * pct, 2)) + "%",
+ "-",
+ ]
+ metrics["Treynor Ratio"] = [
+ str(
+ round(
+ _stats.treynor_ratio(
+ df["returns"], df["benchmark"], win_year, rf
+ )
+ * pct,
+ 2,
+ )
+ )
+ + "%",
+ "-",
+ ]
elif isinstance(returns, _pd.DataFrame):
- greeks = ([_stats.greeks(df[strategy_col], df['benchmark'], win_year, prepare_returns=False)
- for strategy_col in df_strategy_columns])
- metrics['Beta'] = [str(round(g['beta'], 2)) for g in greeks] + ['-']
- metrics['Alpha'] = [str(round(g['alpha'], 2)) for g in greeks] + ['-']
- metrics['Correlation'] = ([str(round(df['benchmark'].corr(df[strategy_col]) * pct, 2)) + '%'
- for strategy_col in df_strategy_columns]) + ['-']
- metrics['Treynor Ratio'] = ([str(round(_stats.treynor_ratio(df[strategy_col], df['benchmark'], win_year, rf) * pct, 2)) + '%'
- for strategy_col in df_strategy_columns]) + ['-']
+ greeks = [
+ _stats.greeks(
+ df[strategy_col],
+ df["benchmark"],
+ win_year,
+ prepare_returns=False,
+ )
+ for strategy_col in df_strategy_columns
+ ]
+ metrics["Beta"] = [str(round(g["beta"], 2)) for g in greeks] + ["-"]
+ metrics["Alpha"] = [str(round(g["alpha"], 2)) for g in greeks] + ["-"]
+ metrics["Correlation"] = (
+ [
+ str(round(df["benchmark"].corr(df[strategy_col]) * pct, 2))
+ + "%"
+ for strategy_col in df_strategy_columns
+ ]
+ ) + ["-"]
+ metrics["Treynor Ratio"] = (
+ [
+ str(
+ round(
+ _stats.treynor_ratio(
+ df[strategy_col], df["benchmark"], win_year, rf
+ )
+ * pct,
+ 2,
+ )
+ )
+ + "%"
+ for strategy_col in df_strategy_columns
+ ]
+ ) + ["-"]
# prepare for display
for col in metrics.columns:
@@ -795,32 +1116,31 @@ def metrics(returns, benchmark=None, rf=0., display=True,
except Exception:
pass
if (display or "internal" in kwargs) and "*int" in col:
- metrics[col] = metrics[col].str.replace('.0', '', regex=False)
+ metrics[col] = metrics[col].str.replace(".0", "", regex=False)
metrics.rename({col: col.replace("*int", "")}, axis=1, inplace=True)
if (display or "internal" in kwargs) and "%" in col:
- metrics[col] = metrics[col] + '%'
+ metrics[col] = metrics[col] + "%"
try:
- metrics['Longest DD Days'] = _pd.to_numeric(
- metrics['Longest DD Days']).astype('int')
- metrics['Avg. Drawdown Days'] = _pd.to_numeric(
- metrics['Avg. Drawdown Days']).astype('int')
+ metrics["Longest DD Days"] = _pd.to_numeric(metrics["Longest DD Days"]).astype(
+ "int"
+ )
+ metrics["Avg. Drawdown Days"] = _pd.to_numeric(
+ metrics["Avg. Drawdown Days"]
+ ).astype("int")
if display or "internal" in kwargs:
- metrics['Longest DD Days'] = metrics['Longest DD Days'].astype(str)
- metrics['Avg. Drawdown Days'] = metrics['Avg. Drawdown Days'
- ].astype(str)
+ metrics["Longest DD Days"] = metrics["Longest DD Days"].astype(str)
+ metrics["Avg. Drawdown Days"] = metrics["Avg. Drawdown Days"].astype(str)
except Exception:
- metrics['Longest DD Days'] = '-'
- metrics['Avg. Drawdown Days'] = '-'
+ metrics["Longest DD Days"] = "-"
+ metrics["Avg. Drawdown Days"] = "-"
if display or "internal" in kwargs:
- metrics['Longest DD Days'] = '-'
- metrics['Avg. Drawdown Days'] = '-'
+ metrics["Longest DD Days"] = "-"
+ metrics["Avg. Drawdown Days"] = "-"
- metrics.columns = [
- col if '~' not in col else '' for col in metrics.columns]
- metrics.columns = [
- col[:-1] if '%' in col else col for col in metrics.columns]
+ metrics.columns = [col if "~" not in col else "" for col in metrics.columns]
+ metrics.columns = [col[:-1] if "%" in col else col for col in metrics.columns]
metrics = metrics.T
if "benchmark" in df:
@@ -836,38 +1156,66 @@ def metrics(returns, benchmark=None, rf=0., display=True,
metrics.columns = [strategy_colname]
# cleanups
- metrics.replace([-0, '-0'], 0, inplace=True)
- metrics.replace([_np.nan, -_np.nan, _np.inf, -_np.inf,
- '-nan%', 'nan%', '-nan', 'nan',
- '-inf%', 'inf%', '-inf', 'inf'], '-', inplace=True)
+ metrics.replace([-0, "-0"], 0, inplace=True)
+ metrics.replace(
+ [
+ _np.nan,
+ -_np.nan,
+ _np.inf,
+ -_np.inf,
+ "-nan%",
+ "nan%",
+ "-nan",
+ "nan",
+ "-inf%",
+ "inf%",
+ "-inf",
+ "inf",
+ ],
+ "-",
+ inplace=True,
+ )
# move benchmark to be the first column always if present
if "benchmark" in df:
- metrics = metrics[[benchmark_colname] + [col for col in metrics.columns if col != benchmark_colname]]
+ metrics = metrics[
+ [benchmark_colname]
+ + [col for col in metrics.columns if col != benchmark_colname]
+ ]
if display:
- print(_tabulate(metrics, headers="keys", tablefmt='simple'))
+ print(_tabulate(metrics, headers="keys", tablefmt="simple"))
return None
if not sep:
- metrics = metrics[metrics.index != '']
+ metrics = metrics[metrics.index != ""]
# remove spaces from column names
metrics = metrics.T
- metrics.columns = [c.replace(' %', '').replace(' *int', '').strip() for c in metrics.columns]
+ metrics.columns = [
+ c.replace(" %", "").replace(" *int", "").strip() for c in metrics.columns
+ ]
metrics = metrics.T
return metrics
-def plots(returns, benchmark=None, grayscale=False,
- figsize=(8, 5), mode='basic', compounded=True,
- periods_per_year=252, prepare_returns=True,
- match_dates=True, **kwargs):
+def plots(
+ returns,
+ benchmark=None,
+ grayscale=False,
+ figsize=(8, 5),
+ mode="basic",
+ compounded=True,
+ periods_per_year=252,
+ prepare_returns=True,
+ match_dates=True,
+ **kwargs,
+):
benchmark_colname = kwargs.get("benchmark_title", "Benchmark")
strategy_colname = kwargs.get("strategy_title", "Strategy")
- active = kwargs.get('active', 'False')
+ active = kwargs.get("active", "False")
if isinstance(returns, _pd.DataFrame):
if len(returns.columns) > 1:
@@ -887,23 +1235,41 @@ def plots(returns, benchmark=None, grayscale=False,
elif isinstance(returns, _pd.DataFrame):
returns.columns = strategy_colname
- if mode.lower() != 'full':
- _plots.snapshot(returns, grayscale=grayscale,
- figsize=(figsize[0], figsize[0]),
- show=True, mode=("comp" if compounded else "sum"),
- benchmark_title=benchmark_colname, strategy_title=strategy_colname)
+ if mode.lower() != "full":
+ _plots.snapshot(
+ returns,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0]),
+ show=True,
+ mode=("comp" if compounded else "sum"),
+ benchmark_title=benchmark_colname,
+ strategy_title=strategy_colname,
+ )
if isinstance(returns, _pd.Series):
- _plots.monthly_heatmap(returns, benchmark, grayscale=grayscale,
- figsize=(figsize[0], figsize[0]*.5),
- show=True, ylabel=False,
- compounded=compounded, active=active)
+ _plots.monthly_heatmap(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ ylabel=False,
+ compounded=compounded,
+ active=active,
+ )
elif isinstance(returns, _pd.DataFrame):
for col in returns.columns:
- _plots.monthly_heatmap(returns[col].dropna(), benchmark, grayscale=grayscale,
- figsize=(figsize[0], figsize[0] * .5),
- show=True, ylabel=False, returns_label=col,
- compounded=compounded, active=active)
+ _plots.monthly_heatmap(
+ returns[col].dropna(),
+ benchmark,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ ylabel=False,
+ returns_label=col,
+ compounded=compounded,
+ active=active,
+ )
return
@@ -916,100 +1282,192 @@ def plots(returns, benchmark=None, grayscale=False,
if match_dates is True:
returns, benchmark = _match_dates(returns, benchmark)
- _plots.returns(returns, benchmark, grayscale=grayscale,
- figsize=(figsize[0], figsize[0]*.6),
- show=True, ylabel=False,
- prepare_returns=False)
+ _plots.returns(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.6),
+ show=True,
+ ylabel=False,
+ prepare_returns=False,
+ )
- _plots.log_returns(returns, benchmark, grayscale=grayscale,
- figsize=(figsize[0], figsize[0]*.5),
- show=True, ylabel=False,
- prepare_returns=False)
+ _plots.log_returns(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ ylabel=False,
+ prepare_returns=False,
+ )
if benchmark is not None:
- _plots.returns(returns, benchmark, match_volatility=True,
- grayscale=grayscale,
- figsize=(figsize[0], figsize[0]*.5),
- show=True, ylabel=False,
- prepare_returns=False)
-
- _plots.yearly_returns(returns, benchmark,
- grayscale=grayscale,
- figsize=(figsize[0], figsize[0]*.5),
- show=True, ylabel=False,
- prepare_returns=False)
-
- _plots.histogram(returns, benchmark, grayscale=grayscale,
- figsize=(figsize[0], figsize[0]*.5),
- show=True, ylabel=False,
- prepare_returns=False)
-
- small_fig_size = (figsize[0], figsize[0]*.35)
- if len(returns.columns) > 1:
- small_fig_size = (figsize[0], figsize[0]*(.33*(len(returns.columns)*.66)))
+ _plots.returns(
+ returns,
+ benchmark,
+ match_volatility=True,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ ylabel=False,
+ prepare_returns=False,
+ )
+
+ _plots.yearly_returns(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ ylabel=False,
+ prepare_returns=False,
+ )
- _plots.daily_returns(returns, benchmark, grayscale=grayscale,
- figsize=small_fig_size,
- show=True, ylabel=False,
- prepare_returns=False, active=active)
+ _plots.histogram(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ ylabel=False,
+ prepare_returns=False,
+ )
- if benchmark is not None:
- _plots.rolling_beta(returns, benchmark, grayscale=grayscale,
- window1=win_half_year, window2=win_year,
- figsize=small_fig_size,
- show=True, ylabel=False,
- prepare_returns=False)
+ small_fig_size = (figsize[0], figsize[0] * 0.35)
+ if len(returns.columns) > 1:
+ small_fig_size = (
+ figsize[0],
+ figsize[0] * (0.33 * (len(returns.columns) * 0.66)),
+ )
+
+ _plots.daily_returns(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=small_fig_size,
+ show=True,
+ ylabel=False,
+ prepare_returns=False,
+ active=active,
+ )
- _plots.rolling_volatility(returns, benchmark, grayscale=grayscale,
- figsize=small_fig_size, show=True, ylabel=False,
- period=win_half_year)
+ if benchmark is not None:
+ _plots.rolling_beta(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ window1=win_half_year,
+ window2=win_year,
+ figsize=small_fig_size,
+ show=True,
+ ylabel=False,
+ prepare_returns=False,
+ )
+
+ _plots.rolling_volatility(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=small_fig_size,
+ show=True,
+ ylabel=False,
+ period=win_half_year,
+ )
- _plots.rolling_sharpe(returns, grayscale=grayscale,
- figsize=small_fig_size,
- show=True, ylabel=False, period=win_half_year)
+ _plots.rolling_sharpe(
+ returns,
+ grayscale=grayscale,
+ figsize=small_fig_size,
+ show=True,
+ ylabel=False,
+ period=win_half_year,
+ )
- _plots.rolling_sortino(returns, grayscale=grayscale,
- figsize=small_fig_size,
- show=True, ylabel=False, period=win_half_year)
+ _plots.rolling_sortino(
+ returns,
+ grayscale=grayscale,
+ figsize=small_fig_size,
+ show=True,
+ ylabel=False,
+ period=win_half_year,
+ )
if isinstance(returns, _pd.Series):
- _plots.drawdowns_periods(returns, grayscale=grayscale,
- figsize=(figsize[0], figsize[0]*.5),
- show=True, ylabel=False,
- prepare_returns=False)
+ _plots.drawdowns_periods(
+ returns,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ ylabel=False,
+ prepare_returns=False,
+ )
elif isinstance(returns, _pd.DataFrame):
for col in returns.columns:
- _plots.drawdowns_periods(returns[col], grayscale=grayscale,
- figsize=(figsize[0], figsize[0] * .5),
- show=True, ylabel=False, title=col,
- prepare_returns=False)
-
- _plots.drawdown(returns, grayscale=grayscale,
- figsize=(figsize[0], figsize[0]*.4),
- show=True, ylabel=False)
+ _plots.drawdowns_periods(
+ returns[col],
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ ylabel=False,
+ title=col,
+ prepare_returns=False,
+ )
+
+ _plots.drawdown(
+ returns,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.4),
+ show=True,
+ ylabel=False,
+ )
if isinstance(returns, _pd.Series):
- _plots.monthly_heatmap(returns, benchmark, grayscale=grayscale,
- figsize=(figsize[0], figsize[0]*.5), returns_label=returns.name,
- show=True, ylabel=False, active=active)
+ _plots.monthly_heatmap(
+ returns,
+ benchmark,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ returns_label=returns.name,
+ show=True,
+ ylabel=False,
+ active=active,
+ )
elif isinstance(returns, _pd.DataFrame):
for col in returns.columns:
- _plots.monthly_heatmap(returns[col], benchmark, grayscale=grayscale,
- figsize=(figsize[0], figsize[0] * .5),
- show=True, ylabel=False, returns_label=col,
- compounded=compounded, active=active)
+ _plots.monthly_heatmap(
+ returns[col],
+ benchmark,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ ylabel=False,
+ returns_label=col,
+ compounded=compounded,
+ active=active,
+ )
if isinstance(returns, _pd.Series):
- _plots.distribution(returns, grayscale=grayscale,
- figsize=(figsize[0], figsize[0] * .5),
- show=True, title=returns.name, ylabel=False,
- prepare_returns=False)
+ _plots.distribution(
+ returns,
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ title=returns.name,
+ ylabel=False,
+ prepare_returns=False,
+ )
elif isinstance(returns, _pd.DataFrame):
for col in returns.columns:
- _plots.distribution(returns[col], grayscale=grayscale,
- figsize=(figsize[0], figsize[0] * .5),
- show=True, title=col, ylabel=False,
- prepare_returns=False)
+ _plots.distribution(
+ returns[col],
+ grayscale=grayscale,
+ figsize=(figsize[0], figsize[0] * 0.5),
+ show=True,
+ title=col,
+ ylabel=False,
+ prepare_returns=False,
+ )
def _calc_dd(df, display=True, as_pct=False):
@@ -1020,96 +1478,127 @@ def _calc_dd(df, display=True, as_pct=False):
return _pd.DataFrame()
if "returns" in dd_info:
- ret_dd = dd_info['returns']
+ ret_dd = dd_info["returns"]
# to match multiple columns like returns_1, returns_2, ...
- elif any(dd_info.columns.get_level_values(0).str.contains('returns')) \
- and dd_info.columns.get_level_values(0).nunique() > 1:
- ret_dd = dd_info.loc[:, dd_info.columns.get_level_values(0).str.contains('returns')]
+ elif (
+ any(dd_info.columns.get_level_values(0).str.contains("returns"))
+ and dd_info.columns.get_level_values(0).nunique() > 1
+ ):
+ ret_dd = dd_info.loc[
+ :, dd_info.columns.get_level_values(0).str.contains("returns")
+ ]
else:
ret_dd = dd_info
- if any(ret_dd.columns.get_level_values(0).str.contains('returns')) and \
- ret_dd.columns.get_level_values(0).nunique() > 1:
+ if (
+ any(ret_dd.columns.get_level_values(0).str.contains("returns"))
+ and ret_dd.columns.get_level_values(0).nunique() > 1
+ ):
dd_stats = {
col: {
- 'Max Drawdown %': ret_dd[col].sort_values(
- by='max drawdown', ascending=True
- )['max drawdown'].values[0] / 100,
- 'Longest DD Days': str(_np.round(ret_dd[col].sort_values(
- by='days', ascending=False)['days'].values[0])),
- 'Avg. Drawdown %': ret_dd[col]['max drawdown'].mean() / 100,
- 'Avg. Drawdown Days': str(_np.round(ret_dd[col]['days'].mean()))
+ "Max Drawdown %": ret_dd[col]
+ .sort_values(by="max drawdown", ascending=True)["max drawdown"]
+ .values[0]
+ / 100,
+ "Longest DD Days": str(
+ _np.round(
+ ret_dd[col]
+ .sort_values(by="days", ascending=False)["days"]
+ .values[0]
+ )
+ ),
+ "Avg. Drawdown %": ret_dd[col]["max drawdown"].mean() / 100,
+ "Avg. Drawdown Days": str(_np.round(ret_dd[col]["days"].mean())),
}
for col in ret_dd.columns.get_level_values(0)
}
else:
dd_stats = {
- 'returns': {
- 'Max Drawdown %': ret_dd.sort_values(
- by='max drawdown', ascending=True
- )['max drawdown'].values[0] / 100,
- 'Longest DD Days': str(_np.round(ret_dd.sort_values(
- by='days', ascending=False)['days'].values[0])),
- 'Avg. Drawdown %': ret_dd['max drawdown'].mean() / 100,
- 'Avg. Drawdown Days': str(_np.round(ret_dd['days'].mean()))
+ "returns": {
+ "Max Drawdown %": ret_dd.sort_values(by="max drawdown", ascending=True)[
+ "max drawdown"
+ ].values[0]
+ / 100,
+ "Longest DD Days": str(
+ _np.round(
+ ret_dd.sort_values(by="days", ascending=False)["days"].values[0]
+ )
+ ),
+ "Avg. Drawdown %": ret_dd["max drawdown"].mean() / 100,
+ "Avg. Drawdown Days": str(_np.round(ret_dd["days"].mean())),
}
}
if "benchmark" in df and (dd_info.columns, _pd.MultiIndex):
- bench_dd = dd_info['benchmark'].sort_values(by='max drawdown')
- dd_stats['benchmark'] = {
- 'Max Drawdown %': bench_dd.sort_values(
- by='max drawdown', ascending=True
- )['max drawdown'].values[0] / 100,
- 'Longest DD Days': str(_np.round(bench_dd.sort_values(
- by='days', ascending=False)['days'].values[0])),
- 'Avg. Drawdown %': bench_dd['max drawdown'].mean() / 100,
- 'Avg. Drawdown Days': str(_np.round(bench_dd['days'].mean()))
+ bench_dd = dd_info["benchmark"].sort_values(by="max drawdown")
+ dd_stats["benchmark"] = {
+ "Max Drawdown %": bench_dd.sort_values(by="max drawdown", ascending=True)[
+ "max drawdown"
+ ].values[0]
+ / 100,
+ "Longest DD Days": str(
+ _np.round(
+ bench_dd.sort_values(by="days", ascending=False)["days"].values[0]
+ )
+ ),
+ "Avg. Drawdown %": bench_dd["max drawdown"].mean() / 100,
+ "Avg. Drawdown Days": str(_np.round(bench_dd["days"].mean())),
}
# pct multiplier
pct = 100 if display or as_pct else 1
dd_stats = _pd.DataFrame(dd_stats).T
- dd_stats['Max Drawdown %'] = dd_stats['Max Drawdown %'].astype(float) * pct
- dd_stats['Avg. Drawdown %'] = dd_stats['Avg. Drawdown %'].astype(float) * pct
+ dd_stats["Max Drawdown %"] = dd_stats["Max Drawdown %"].astype(float) * pct
+ dd_stats["Avg. Drawdown %"] = dd_stats["Avg. Drawdown %"].astype(float) * pct
return dd_stats.T
def _html_table(obj, showindex="default"):
- obj = _tabulate(obj, headers="keys", tablefmt='html',
- floatfmt=".2f", showindex=showindex)
- obj = obj.replace(' style="text-align: right;"', '')
- obj = obj.replace(' style="text-align: left;"', '')
- obj = obj.replace(' style="text-align: center;"', '')
- obj = _regex.sub(' +', ' | ', obj)
- obj = _regex.sub(' + | ', '', obj)
- obj = _regex.sub(' +', ' | ', obj)
- obj = _regex.sub(' + | ', '', obj)
+ obj = _tabulate(
+ obj, headers="keys", tablefmt="html", floatfmt=".2f", showindex=showindex
+ )
+ obj = obj.replace(' style="text-align: right;"', "")
+ obj = obj.replace(' style="text-align: left;"', "")
+ obj = obj.replace(' style="text-align: center;"', "")
+ obj = _regex.sub(" +", " | ", obj)
+ obj = _regex.sub(" + | ", "", obj)
+ obj = _regex.sub(" +", " | ", obj)
+ obj = _regex.sub(" + | ", "", obj)
return obj
def _download_html(html, filename="quantstats-tearsheet.html"):
- jscode = _regex.sub(' +', ' ', """""".replace('\n', ''))
- jscode = jscode.replace('{{html}}', _regex.sub(
- ' +', ' ', html.replace('\n', '')))
+ a.click();""".replace(
+ "\n", ""
+ ),
+ )
+ jscode = jscode.replace("{{html}}", _regex.sub(" +", " ", html.replace("\n", "")))
if _utils._in_notebook():
- iDisplay(iHTML(jscode.replace('{{filename}}', filename)))
+ iDisplay(iHTML(jscode.replace("{{filename}}", filename)))
def _open_html(html):
- jscode = _regex.sub(' +', ' ', """""".replace('\n', ''))
- jscode = jscode.replace('{{html}}', _regex.sub(
- ' +', ' ', html.replace('\n', '')))
+ """.replace(
+ "\n", ""
+ ),
+ )
+ jscode = jscode.replace("{{html}}", _regex.sub(" +", " ", html.replace("\n", "")))
if _utils._in_notebook():
iDisplay(iHTML(jscode))
@@ -1119,13 +1608,15 @@ def _embed_figure(figfiles, figfmt):
embed_string = "\n"
for figfile in figfiles:
figbytes = figfile.getvalue()
- if figfmt == 'svg':
+ if figfmt == "svg":
return figbytes.decode()
data_uri = _b64encode(figbytes).decode()
- embed_string.join('
'.format(figfmt, data_uri))
+ embed_string.join(
+ ''.format(figfmt, data_uri)
+ )
else:
figbytes = figfiles.getvalue()
- if figfmt == 'svg':
+ if figfmt == "svg":
return figbytes.decode()
data_uri = _b64encode(figbytes).decode()
embed_string = ''.format(figfmt, data_uri)
diff --git a/quantstats/stats.py b/quantstats/stats.py
index 7e81158e..27346f61 100644
--- a/quantstats/stats.py
+++ b/quantstats/stats.py
@@ -22,19 +22,18 @@
import pandas as _pd
import numpy as _np
from math import ceil as _ceil, sqrt as _sqrt
-from scipy.stats import (
- norm as _norm, linregress as _linregress
-)
+from scipy.stats import norm as _norm, linregress as _linregress
from . import utils as _utils
# ======== STATS ========
+
def pct_rank(prices, window=60):
"""Rank prices by window"""
rank = _utils.multi_shift(prices, window).T.rank(pct=True).T
- return rank.iloc[:, 0] * 100.
+ return rank.iloc[:, 0] * 100.0
def compsum(returns):
@@ -60,12 +59,14 @@ def get_outliers(data):
}
if isinstance(returns, _pd.DataFrame):
- warn("Pandas DataFrame was passed (Series expected). "
- "Only first column will be used.")
+ warn(
+ "Pandas DataFrame was passed (Series expected). "
+ "Only first column will be used."
+ )
returns = returns.copy()
returns.columns = map(str.lower, returns.columns)
- if len(returns.columns) > 1 and 'close' in returns.columns:
- returns = returns['close']
+ if len(returns.columns) > 1 and "close" in returns.columns:
+ returns = returns["close"]
else:
returns = returns[returns.columns[0]]
@@ -77,15 +78,14 @@ def get_outliers(data):
return {
"Daily": get_outliers(daily),
- "Weekly": get_outliers(daily.resample('W-MON').apply(apply_fnc)),
- "Monthly": get_outliers(daily.resample('M').apply(apply_fnc)),
- "Quarterly": get_outliers(daily.resample('Q').apply(apply_fnc)),
- "Yearly": get_outliers(daily.resample('A').apply(apply_fnc))
+ "Weekly": get_outliers(daily.resample("W-MON").apply(apply_fnc)),
+ "Monthly": get_outliers(daily.resample("M").apply(apply_fnc)),
+ "Quarterly": get_outliers(daily.resample("Q").apply(apply_fnc)),
+ "Yearly": get_outliers(daily.resample("A").apply(apply_fnc)),
}
-def expected_return(returns, aggregate=None, compounded=True,
- prepare_returns=True):
+def expected_return(returns, aggregate=None, compounded=True, prepare_returns=True):
"""
Returns the expected return for a given period
by calculating the geometric holding period return
@@ -106,12 +106,12 @@ def ghpr(retruns, aggregate=None, compounded=True):
return expected_return(retruns, aggregate, compounded)
-def outliers(returns, quantile=.95):
+def outliers(returns, quantile=0.95):
"""Returns series of outliers"""
- return returns[returns > returns.quantile(quantile)].dropna(how='all')
+ return returns[returns > returns.quantile(quantile)].dropna(how="all")
-def remove_outliers(returns, quantile=.95):
+def remove_outliers(returns, quantile=0.95):
"""Returns series of returns without the outliers"""
return returns[returns < returns.quantile(quantile)]
@@ -130,8 +130,7 @@ def worst(returns, aggregate=None, compounded=True, prepare_returns=True):
return _utils.aggregate_returns(returns, aggregate, compounded).min()
-def consecutive_wins(returns, aggregate=None, compounded=True,
- prepare_returns=True):
+def consecutive_wins(returns, aggregate=None, compounded=True, prepare_returns=True):
"""Returns the maximum consecutive wins by day/month/week/quarter/year"""
if prepare_returns:
returns = _utils._prepare_returns(returns)
@@ -139,8 +138,7 @@ def consecutive_wins(returns, aggregate=None, compounded=True,
return _utils._count_consecutive(returns).max()
-def consecutive_losses(returns, aggregate=None, compounded=True,
- prepare_returns=True):
+def consecutive_losses(returns, aggregate=None, compounded=True, prepare_returns=True):
"""
Returns the maximum consecutive losses by
day/month/week/quarter/year
@@ -170,11 +168,12 @@ def _exposure(ret):
def win_rate(returns, aggregate=None, compounded=True, prepare_returns=True):
"""Calculates the win ratio for a period"""
+
def _win_rate(series):
try:
return len(series[series > 0]) / len(series[series != 0])
except Exception:
- return 0.
+ return 0.0
if prepare_returns:
returns = _utils._prepare_returns(returns)
@@ -235,8 +234,9 @@ def volatility(returns, periods=252, annualize=True, prepare_returns=True):
return std
-def rolling_volatility(returns, rolling_period=126, periods_per_year=252,
- prepare_returns=True):
+def rolling_volatility(
+ returns, rolling_period=126, periods_per_year=252, prepare_returns=True
+):
if prepare_returns:
returns = _utils._prepare_returns(returns, rolling_period)
@@ -262,13 +262,14 @@ def autocorr_penalty(returns, prepare_returns=False):
# returns.to_csv('/Users/ran/Desktop/test.csv')
num = len(returns)
coef = _np.abs(_np.corrcoef(returns[:-1], returns[1:])[0, 1])
- corr = [((num - x)/num) * coef ** x for x in range(1, num)]
+ corr = [((num - x) / num) * coef**x for x in range(1, num)]
return _np.sqrt(1 + 2 * _np.sum(corr))
# ======= METRICS =======
-def sharpe(returns, rf=0., periods=252, annualize=True, smart=False):
+
+def sharpe(returns, rf=0.0, periods=252, annualize=True, smart=False):
"""
Calculates the sharpe ratio of access returns
@@ -283,7 +284,7 @@ def sharpe(returns, rf=0., periods=252, annualize=True, smart=False):
* smart: return smart sharpe ratio
"""
if rf != 0 and periods is None:
- raise Exception('Must provide periods if rf != 0')
+ raise Exception("Must provide periods if rf != 0")
returns = _utils._prepare_returns(returns, rf, periods)
divisor = returns.std(ddof=1)
@@ -293,32 +294,34 @@ def sharpe(returns, rf=0., periods=252, annualize=True, smart=False):
res = returns.mean() / divisor
if annualize:
- return res * _np.sqrt(
- 1 if periods is None else periods)
+ return res * _np.sqrt(1 if periods is None else periods)
return res
-def smart_sharpe(returns, rf=0., periods=252, annualize=True):
+def smart_sharpe(returns, rf=0.0, periods=252, annualize=True):
return sharpe(returns, rf, periods, annualize, True)
-def rolling_sharpe(returns, rf=0., rolling_period=126,
- annualize=True, periods_per_year=252,
- prepare_returns=True):
+def rolling_sharpe(
+ returns,
+ rf=0.0,
+ rolling_period=126,
+ annualize=True,
+ periods_per_year=252,
+ prepare_returns=True,
+):
if rf != 0 and rolling_period is None:
- raise Exception('Must provide periods if rf != 0')
+ raise Exception("Must provide periods if rf != 0")
if prepare_returns:
returns = _utils._prepare_returns(returns, rf, rolling_period)
- res = returns.rolling(rolling_period).mean() / \
- returns.rolling(rolling_period).std()
+ res = returns.rolling(rolling_period).mean() / returns.rolling(rolling_period).std()
if annualize:
- res = res * _np.sqrt(
- 1 if periods_per_year is None else periods_per_year)
+ res = res * _np.sqrt(1 if periods_per_year is None else periods_per_year)
return res
@@ -333,7 +336,7 @@ def sortino(returns, rf=0, periods=252, annualize=True, smart=False):
http://www.redrockcapital.com/Sortino__A__Sharper__Ratio_Red_Rock_Capital.pdf
"""
if rf != 0 and periods is None:
- raise Exception('Must provide periods if rf != 0')
+ raise Exception("Must provide periods if rf != 0")
returns = _utils._prepare_returns(returns, rf, periods)
@@ -346,8 +349,7 @@ def sortino(returns, rf=0, periods=252, annualize=True, smart=False):
res = returns.mean() / downside
if annualize:
- return res * _np.sqrt(
- 1 if periods is None else periods)
+ return res * _np.sqrt(1 if periods is None else periods)
return res
@@ -356,21 +358,25 @@ def smart_sortino(returns, rf=0, periods=252, annualize=True):
return sortino(returns, rf, periods, annualize, True)
-def rolling_sortino(returns, rf=0, rolling_period=126, annualize=True,
- periods_per_year=252, **kwargs):
+def rolling_sortino(
+ returns, rf=0, rolling_period=126, annualize=True, periods_per_year=252, **kwargs
+):
if rf != 0 and rolling_period is None:
- raise Exception('Must provide periods if rf != 0')
+ raise Exception("Must provide periods if rf != 0")
if kwargs.get("prepare_returns", True):
returns = _utils._prepare_returns(returns, rf, rolling_period)
- downside = returns.rolling(rolling_period).apply(
- lambda x: (x.values[x.values < 0]**2).sum()) / rolling_period
+ downside = (
+ returns.rolling(rolling_period).apply(
+ lambda x: (x.values[x.values < 0] ** 2).sum()
+ )
+ / rolling_period
+ )
res = returns.rolling(rolling_period).mean() / _np.sqrt(downside)
if annualize:
- res = res * _np.sqrt(
- 1 if periods_per_year is None else periods_per_year)
+ res = res * _np.sqrt(1 if periods_per_year is None else periods_per_year)
return res
@@ -380,55 +386,77 @@ def adjusted_sortino(returns, rf=0, periods=252, annualize=True, smart=False):
direct comparisons to the Sharpe. See here for more info:
https://archive.is/wip/2rwFW
"""
- data = sortino(
- returns, rf, periods=periods, annualize=annualize, smart=smart)
+ data = sortino(returns, rf, periods=periods, annualize=annualize, smart=smart)
return data / _sqrt(2)
-def probabilistic_ratio(series, rf=0., base="sharpe", periods=252, annualize=False, smart=False):
+def probabilistic_ratio(
+ series, rf=0.0, base="sharpe", periods=252, annualize=False, smart=False
+):
if base.lower() == "sharpe":
base = sharpe(series, periods=periods, annualize=False, smart=smart)
elif base.lower() == "sortino":
base = sortino(series, periods=periods, annualize=False, smart=smart)
elif base.lower() == "adjusted_sortino":
- base = adjusted_sortino(series, periods=periods,
- annualize=False, smart=smart)
+ base = adjusted_sortino(series, periods=periods, annualize=False, smart=smart)
else:
raise Exception(
- '`metric` must be either `sharpe`, `sortino`, or `adjusted_sortino`')
+ "`metric` must be either `sharpe`, `sortino`, or `adjusted_sortino`"
+ )
skew_no = skew(series, prepare_returns=False)
kurtosis_no = kurtosis(series, prepare_returns=False)
n = len(series)
- sigma_sr = _np.sqrt((1 + (0.5 * base ** 2) - (skew_no * base) +
- (((kurtosis_no - 3) / 4) * base ** 2)) / (n - 1))
+ sigma_sr = _np.sqrt(
+ (
+ 1
+ + (0.5 * base**2)
+ - (skew_no * base)
+ + (((kurtosis_no - 3) / 4) * base**2)
+ )
+ / (n - 1)
+ )
ratio = (base - rf) / sigma_sr
psr = _norm.cdf(ratio)
if annualize:
- return psr * (252 ** 0.5)
+ return psr * (252**0.5)
return psr
-def probabilistic_sharpe_ratio(series, rf=0., periods=252, annualize=False, smart=False):
- return probabilistic_ratio(series, rf, base="sharpe", periods=periods,
- annualize=annualize, smart=smart)
+def probabilistic_sharpe_ratio(
+ series, rf=0.0, periods=252, annualize=False, smart=False
+):
+ return probabilistic_ratio(
+ series, rf, base="sharpe", periods=periods, annualize=annualize, smart=smart
+ )
-def probabilistic_sortino_ratio(series, rf=0., periods=252, annualize=False, smart=False):
- return probabilistic_ratio(series, rf, base="sortino", periods=periods,
- annualize=annualize, smart=smart)
+def probabilistic_sortino_ratio(
+ series, rf=0.0, periods=252, annualize=False, smart=False
+):
+ return probabilistic_ratio(
+ series, rf, base="sortino", periods=periods, annualize=annualize, smart=smart
+ )
-def probabilistic_adjusted_sortino_ratio(series, rf=0., periods=252, annualize=False, smart=False):
- return probabilistic_ratio(series, rf, base="adjusted_sortino", periods=periods,
- annualize=annualize, smart=smart)
+def probabilistic_adjusted_sortino_ratio(
+ series, rf=0.0, periods=252, annualize=False, smart=False
+):
+ return probabilistic_ratio(
+ series,
+ rf,
+ base="adjusted_sortino",
+ periods=periods,
+ annualize=annualize,
+ smart=smart,
+ )
-def treynor_ratio(returns, benchmark, periods=252., rf=0.):
+def treynor_ratio(returns, benchmark, periods=252.0, rf=0.0):
"""
Calculates the Treynor ratio
@@ -440,7 +468,7 @@ def treynor_ratio(returns, benchmark, periods=252., rf=0.):
if isinstance(returns, _pd.DataFrame):
returns = returns[returns.columns[0]]
- beta = greeks(returns, benchmark, periods=periods).to_dict().get('beta', 0)
+ beta = greeks(returns, benchmark, periods=periods).to_dict().get("beta", 0)
if beta == 0:
return 0
return (comp(returns) - rf) / beta
@@ -462,12 +490,11 @@ def omega(returns, rf=0.0, required_return=0.0, periods=252):
if periods == 1:
return_threshold = required_return
else:
- return_threshold = (1 + required_return) ** (1. / periods) - 1
+ return_threshold = (1 + required_return) ** (1.0 / periods) - 1
returns_less_thresh = returns - return_threshold
numer = returns_less_thresh[returns_less_thresh > 0.0].sum().values[0]
- denom = -1.0 * \
- returns_less_thresh[returns_less_thresh < 0.0].sum().values[0]
+ denom = -1.0 * returns_less_thresh[returns_less_thresh < 0.0].sum().values[0]
if denom > 0.0:
return numer / denom
@@ -485,7 +512,7 @@ def gain_to_pain_ratio(returns, rf=0, resolution="D"):
return returns.sum() / downside
-def cagr(returns, rf=0., periods=252 ,compounded=True):
+def cagr(returns, rf=0.0, periods=252, compounded=True):
"""
Calculates the communicative annualized growth return
(CAGR%) of access returns
@@ -510,7 +537,7 @@ def cagr(returns, rf=0., periods=252 ,compounded=True):
return res
-def rar(returns, rf=0.):
+def rar(returns, rf=0.0):
"""
Calculates the risk-adjusted return of access returns
(CAGR / exposure. takes time into account.)
@@ -562,7 +589,7 @@ def ulcer_performance_index(returns, rf=0):
Calculates the ulcer index score
(downside risk measurment)
"""
- return (comp(returns)-rf) / ulcer_index(returns)
+ return (comp(returns) - rf) / ulcer_index(returns)
def upi(returns, rf=0):
@@ -576,8 +603,8 @@ def serenity_index(returns, rf=0):
(https://www.keyquant.com/Download/GetFile?Filename=%5CPublications%5CKeyQuant_WhitePaper_APT_Part1.pdf)
"""
dd = to_drawdown_series(returns)
- pitfall = - cvar(dd) / returns.std()
- return (comp(returns)-rf) / (ulcer_index(returns) * pitfall)
+ pitfall = -cvar(dd) / returns.std()
+ return (comp(returns) - rf) / (ulcer_index(returns) * pitfall)
def risk_of_ruin(returns, prepare_returns=True):
@@ -607,9 +634,9 @@ def value_at_risk(returns, sigma=1, confidence=0.95, prepare_returns=True):
sigma *= returns.std()
if confidence > 1:
- confidence = confidence/100
+ confidence = confidence / 100
- return _norm.ppf(1-confidence, mu, sigma)
+ return _norm.ppf(1 - confidence, mu, sigma)
def var(returns, sigma=1, confidence=0.95, prepare_returns=True):
@@ -617,8 +644,7 @@ def var(returns, sigma=1, confidence=0.95, prepare_returns=True):
return value_at_risk(returns, sigma, confidence, prepare_returns)
-def conditional_value_at_risk(returns, sigma=1, confidence=0.95,
- prepare_returns=True):
+def conditional_value_at_risk(returns, sigma=1, confidence=0.95, prepare_returns=True):
"""
Calculats the conditional daily value-at-risk (aka expected shortfall)
quantifies the amount of tail risk an investment
@@ -632,8 +658,7 @@ def conditional_value_at_risk(returns, sigma=1, confidence=0.95,
def cvar(returns, sigma=1, confidence=0.95, prepare_returns=True):
"""Shorthand for conditional_value_at_risk()"""
- return conditional_value_at_risk(
- returns, sigma, confidence, prepare_returns)
+ return conditional_value_at_risk(returns, sigma, confidence, prepare_returns)
def expected_shortfall(returns, sigma=1, confidence=0.95):
@@ -648,7 +673,7 @@ def tail_ratio(returns, cutoff=0.95, prepare_returns=True):
"""
if prepare_returns:
returns = _utils._prepare_returns(returns)
- return abs(returns.quantile(cutoff) / returns.quantile(1-cutoff))
+ return abs(returns.quantile(cutoff) / returns.quantile(1 - cutoff))
def payoff_ratio(returns, prepare_returns=True):
@@ -675,7 +700,7 @@ def profit_ratio(returns, prepare_returns=True):
try:
return win_ratio / loss_ratio
except Exception:
- return 0.
+ return 0.0
def profit_factor(returns, prepare_returns=True):
@@ -692,8 +717,7 @@ def cpc_index(returns, prepare_returns=True):
"""
if prepare_returns:
returns = _utils._prepare_returns(returns)
- return profit_factor(returns) * win_rate(returns) * \
- win_loss_ratio(returns)
+ return profit_factor(returns) * win_rate(returns) * win_loss_ratio(returns)
def common_sense_ratio(returns, prepare_returns=True):
@@ -703,7 +727,7 @@ def common_sense_ratio(returns, prepare_returns=True):
return profit_factor(returns) * tail_ratio(returns)
-def outlier_win_ratio(returns, quantile=.99, prepare_returns=True):
+def outlier_win_ratio(returns, quantile=0.99, prepare_returns=True):
"""
Calculates the outlier winners ratio
99th percentile of returns / mean positive return
@@ -713,7 +737,7 @@ def outlier_win_ratio(returns, quantile=.99, prepare_returns=True):
return returns.quantile(quantile).mean() / returns[returns >= 0].mean()
-def outlier_loss_ratio(returns, quantile=.01, prepare_returns=True):
+def outlier_loss_ratio(returns, quantile=0.01, prepare_returns=True):
"""
Calculates the outlier losers ratio
1st percentile of returns / mean negative return
@@ -751,7 +775,7 @@ def max_drawdown(prices):
def to_drawdown_series(returns):
"""Convert returns series to drawdown series"""
prices = _utils._prepare_prices(returns)
- dd = prices / _np.maximum.accumulate(prices) - 1.
+ dd = prices / _np.maximum.accumulate(prices) - 1.0
return dd.replace([_np.inf, -_np.inf, -0], 0)
@@ -761,6 +785,7 @@ def drawdown_details(drawdown):
duration, max drawdown and max dd for 99% of the dd period
for every drawdown period
"""
+
def _drawdown_details(drawdown):
# mark no drawdown
no_dd = drawdown == 0
@@ -777,8 +802,16 @@ def _drawdown_details(drawdown):
# no drawdown :)
if not starts:
return _pd.DataFrame(
- index=[], columns=('start', 'valley', 'end', 'days',
- 'max drawdown', '99% max drawdown'))
+ index=[],
+ columns=(
+ "start",
+ "valley",
+ "end",
+ "days",
+ "max drawdown",
+ "99% max drawdown",
+ ),
+ )
# drawdown series begins in a drawdown
if ends and starts[0] > ends[0]:
@@ -791,23 +824,37 @@ def _drawdown_details(drawdown):
# build dataframe from results
data = []
for i, _ in enumerate(starts):
- dd = drawdown[starts[i]:ends[i]]
- clean_dd = -remove_outliers(-dd, .99)
- data.append((starts[i], dd.idxmin(), ends[i],
- (ends[i] - starts[i]).days+1,
- dd.min() * 100, clean_dd.min() * 100))
-
- df = _pd.DataFrame(data=data,
- columns=('start', 'valley', 'end', 'days',
- 'max drawdown',
- '99% max drawdown'))
- df['days'] = df['days'].astype(int)
- df['max drawdown'] = df['max drawdown'].astype(float)
- df['99% max drawdown'] = df['99% max drawdown'].astype(float)
-
- df['start'] = df['start'].dt.strftime('%Y-%m-%d')
- df['end'] = df['end'].dt.strftime('%Y-%m-%d')
- df['valley'] = df['valley'].dt.strftime('%Y-%m-%d')
+ dd = drawdown[starts[i] : ends[i]]
+ clean_dd = -remove_outliers(-dd, 0.99)
+ data.append(
+ (
+ starts[i],
+ dd.idxmin(),
+ ends[i],
+ (ends[i] - starts[i]).days + 1,
+ dd.min() * 100,
+ clean_dd.min() * 100,
+ )
+ )
+
+ df = _pd.DataFrame(
+ data=data,
+ columns=(
+ "start",
+ "valley",
+ "end",
+ "days",
+ "max drawdown",
+ "99% max drawdown",
+ ),
+ )
+ df["days"] = df["days"].astype(int)
+ df["max drawdown"] = df["max drawdown"].astype(float)
+ df["99% max drawdown"] = df["99% max drawdown"].astype(float)
+
+ df["start"] = df["start"].dt.strftime("%Y-%m-%d")
+ df["end"] = df["end"].dt.strftime("%Y-%m-%d")
+ df["valley"] = df["valley"].dt.strftime("%Y-%m-%d")
return df
@@ -837,13 +884,15 @@ def kelly_criterion(returns, prepare_returns=True):
# ==== VS. BENCHMARK ====
+
def r_squared(returns, benchmark, prepare_returns=True):
"""Measures the straight line fit of the equity curve"""
# slope, intercept, r_val, p_val, std_err = _linregress(
if prepare_returns:
returns = _utils._prepare_returns(returns)
_, _, r_val, _, _ = _linregress(
- returns, _utils._prepare_benchmark(benchmark, returns.index))
+ returns, _utils._prepare_benchmark(benchmark, returns.index)
+ )
return r_val**2
@@ -864,7 +913,7 @@ def information_ratio(returns, benchmark, prepare_returns=True):
return diff_rets.mean() / diff_rets.std()
-def greeks(returns, benchmark, periods=252., prepare_returns=True):
+def greeks(returns, benchmark, periods=252.0, prepare_returns=True):
"""Calculates alpha and beta of the portfolio"""
# ----------------------------
# data cleanup
@@ -881,37 +930,44 @@ def greeks(returns, benchmark, periods=252., prepare_returns=True):
alpha = returns.mean() - beta * benchmark.mean()
alpha = alpha * periods
- return _pd.Series({
- "beta": beta,
- "alpha": alpha,
- # "vol": _np.sqrt(matrix[0, 0]) * _np.sqrt(periods)
- }).fillna(0)
+ return _pd.Series(
+ {
+ "beta": beta,
+ "alpha": alpha,
+ # "vol": _np.sqrt(matrix[0, 0]) * _np.sqrt(periods)
+ }
+ ).fillna(0)
def rolling_greeks(returns, benchmark, periods=252, prepare_returns=True):
"""Calculates rolling alpha and beta of the portfolio"""
if prepare_returns:
returns = _utils._prepare_returns(returns)
- df = _pd.DataFrame(data={
- "returns": returns,
- "benchmark": _utils._prepare_benchmark(benchmark, returns.index)
- })
+ df = _pd.DataFrame(
+ data={
+ "returns": returns,
+ "benchmark": _utils._prepare_benchmark(benchmark, returns.index),
+ }
+ )
df = df.fillna(0)
- corr = df.rolling(int(periods)).corr().unstack()['returns']['benchmark']
+ corr = df.rolling(int(periods)).corr().unstack()["returns"]["benchmark"]
std = df.rolling(int(periods)).std()
- beta = corr * std['returns'] / std['benchmark']
+ beta = corr * std["returns"] / std["benchmark"]
- alpha = df['returns'].mean() - beta * df['benchmark'].mean()
+ alpha = df["returns"].mean() - beta * df["benchmark"].mean()
# alpha = alpha * periods
- return _pd.DataFrame(index=returns.index, data={
- "beta": beta,
- "alpha": alpha
- })
+ return _pd.DataFrame(index=returns.index, data={"beta": beta, "alpha": alpha})
-def compare(returns, benchmark, aggregate=None, compounded=True,
- round_vals=None, prepare_returns=True):
+def compare(
+ returns,
+ benchmark,
+ aggregate=None,
+ compounded=True,
+ round_vals=None,
+ prepare_returns=True,
+):
"""
Compare returns to benchmark on a
day/week/month/quarter/year basis
@@ -921,19 +977,28 @@ def compare(returns, benchmark, aggregate=None, compounded=True,
benchmark = _utils._prepare_benchmark(benchmark, returns.index)
if isinstance(returns, _pd.Series):
- data = _pd.DataFrame(data={
- 'Benchmark': _utils.aggregate_returns(
- benchmark, aggregate, compounded) * 100,
- 'Returns': _utils.aggregate_returns(
- returns, aggregate, compounded) * 100
- })
-
- data['Multiplier'] = data['Returns'] / data['Benchmark']
- data['Won'] = _np.where(data['Returns'] >= data['Benchmark'], '+', '-')
+ data = _pd.DataFrame(
+ data={
+ "Benchmark": _utils.aggregate_returns(benchmark, aggregate, compounded)
+ * 100,
+ "Returns": _utils.aggregate_returns(returns, aggregate, compounded)
+ * 100,
+ }
+ )
+
+ data["Multiplier"] = data["Returns"] / data["Benchmark"]
+ data["Won"] = _np.where(data["Returns"] >= data["Benchmark"], "+", "-")
elif isinstance(returns, _pd.DataFrame):
- bench = {'Benchmark': _utils.aggregate_returns(benchmark, aggregate, compounded) * 100}
- strategy = {'Returns_' + str(i): _utils.aggregate_returns(returns[col], aggregate, compounded) * 100
- for i, col in enumerate(returns.columns)}
+ bench = {
+ "Benchmark": _utils.aggregate_returns(benchmark, aggregate, compounded)
+ * 100
+ }
+ strategy = {
+ "Returns_"
+ + str(i): _utils.aggregate_returns(returns[col], aggregate, compounded)
+ * 100
+ for i, col in enumerate(returns.columns)
+ }
data = _pd.DataFrame(data={**bench, **strategy})
if round_vals is not None:
@@ -945,12 +1010,14 @@ def compare(returns, benchmark, aggregate=None, compounded=True,
def monthly_returns(returns, eoy=True, compounded=True, prepare_returns=True):
"""Calculates monthly returns"""
if isinstance(returns, _pd.DataFrame):
- warn("Pandas DataFrame was passed (Series expected). "
- "Only first column will be used.")
+ warn(
+ "Pandas DataFrame was passed (Series expected). "
+ "Only first column will be used."
+ )
returns = returns.copy()
returns.columns = map(str.lower, returns.columns)
- if len(returns.columns) > 1 and 'close' in returns.columns:
- returns = returns['close']
+ if len(returns.columns) > 1 and "close" in returns.columns:
+ returns = returns["close"]
else:
returns = returns[returns.columns[0]]
@@ -959,35 +1026,59 @@ def monthly_returns(returns, eoy=True, compounded=True, prepare_returns=True):
original_returns = returns.copy()
returns = _pd.DataFrame(
- _utils.group_returns(returns,
- returns.index.strftime('%Y-%m-01'),
- compounded))
+ _utils.group_returns(returns, returns.index.strftime("%Y-%m-01"), compounded)
+ )
- returns.columns = ['Returns']
+ returns.columns = ["Returns"]
returns.index = _pd.to_datetime(returns.index)
# get returnsframe
- returns['Year'] = returns.index.strftime('%Y')
- returns['Month'] = returns.index.strftime('%b')
+ returns["Year"] = returns.index.strftime("%Y")
+ returns["Month"] = returns.index.strftime("%b")
# make pivot table
- returns = returns.pivot(index='Year', columns='Month', values='Returns').fillna(0)
+ returns = returns.pivot(index="Year", columns="Month", values="Returns").fillna(0)
# handle missing months
- for month in ['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
- 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']:
+ for month in [
+ "Jan",
+ "Feb",
+ "Mar",
+ "Apr",
+ "May",
+ "Jun",
+ "Jul",
+ "Aug",
+ "Sep",
+ "Oct",
+ "Nov",
+ "Dec",
+ ]:
if month not in returns.columns:
returns.loc[:, month] = 0
# order columns by month
- returns = returns[['Jan', 'Feb', 'Mar', 'Apr', 'May', 'Jun',
- 'Jul', 'Aug', 'Sep', 'Oct', 'Nov', 'Dec']]
+ returns = returns[
+ [
+ "Jan",
+ "Feb",
+ "Mar",
+ "Apr",
+ "May",
+ "Jun",
+ "Jul",
+ "Aug",
+ "Sep",
+ "Oct",
+ "Nov",
+ "Dec",
+ ]
+ ]
if eoy:
- returns['eoy'] = _utils.group_returns(
- original_returns,
- original_returns.index.year,
- compounded=compounded).values
+ returns["eoy"] = _utils.group_returns(
+ original_returns, original_returns.index.year, compounded=compounded
+ ).values
returns.columns = map(lambda x: str(x).upper(), returns.columns)
returns.index.name = None
diff --git a/quantstats/utils.py b/quantstats/utils.py
index e788b391..51eff8db 100644
--- a/quantstats/utils.py
+++ b/quantstats/utils.py
@@ -27,22 +27,19 @@
def _mtd(df):
- return df[df.index >= _dt.datetime.now(
- ).strftime('%Y-%m-01')]
+ return df[df.index >= _dt.datetime.now().strftime("%Y-%m-01")]
def _qtd(df):
date = _dt.datetime.now()
for q in [1, 4, 7, 10]:
if date.month <= q:
- return df[df.index >= _dt.datetime(
- date.year, q, 1).strftime('%Y-%m-01')]
- return df[df.index >= date.strftime('%Y-%m-01')]
+ return df[df.index >= _dt.datetime(date.year, q, 1).strftime("%Y-%m-01")]
+ return df[df.index >= date.strftime("%Y-%m-01")]
def _ytd(df):
- return df[df.index >= _dt.datetime.now(
- ).strftime('%Y-01-01')]
+ return df[df.index >= _dt.datetime.now().strftime("%Y-01-01")]
def _pandas_date(df, dates):
@@ -68,42 +65,42 @@ def multi_shift(df, shift=3):
return _pd.concat(dfs, 1, sort=True)
-def to_returns(prices, rf=0.):
+def to_returns(prices, rf=0.0):
"""Calculates the simple arithmetic returns of a price series"""
return _prepare_returns(prices, rf)
def to_prices(returns, base=1e5):
"""Converts returns series to price data"""
- returns = returns.copy().fillna(0).replace(
- [_np.inf, -_np.inf], float('NaN'))
+ returns = returns.copy().fillna(0).replace([_np.inf, -_np.inf], float("NaN"))
return base + base * _stats.compsum(returns)
-def log_returns(returns, rf=0., nperiods=None):
+def log_returns(returns, rf=0.0, nperiods=None):
"""Shorthand for to_log_returns"""
return to_log_returns(returns, rf, nperiods)
-def to_log_returns(returns, rf=0., nperiods=None):
+def to_log_returns(returns, rf=0.0, nperiods=None):
"""Converts returns series to log returns"""
returns = _prepare_returns(returns, rf, nperiods)
try:
- return _np.log(returns+1).replace([_np.inf, -_np.inf], float('NaN'))
+ return _np.log(returns + 1).replace([_np.inf, -_np.inf], float("NaN"))
except Exception:
- return 0.
+ return 0.0
def exponential_stdev(returns, window=30, is_halflife=False):
"""Returns series representing exponential volatility of returns"""
returns = _prepare_returns(returns)
halflife = window if is_halflife else None
- return returns.ewm(com=None, span=window,
- halflife=halflife, min_periods=window).std()
+ return returns.ewm(
+ com=None, span=window, halflife=halflife, min_periods=window
+ ).std()
-def rebase(prices, base=100.):
+def rebase(prices, base=100.0):
"""
Rebase all series to a given intial base.
This makes comparing/plotting different series together easier.
@@ -126,33 +123,32 @@ def group_returns(returns, groupby, compounded=False):
def aggregate_returns(returns, period=None, compounded=True):
"""Aggregates returns based on date periods"""
- if period is None or 'day' in period:
+ if period is None or "day" in period:
return returns
index = returns.index
- if 'month' in period:
+ if "month" in period:
return group_returns(returns, index.month, compounded=compounded)
- if 'quarter' in period:
+ if "quarter" in period:
return group_returns(returns, index.quarter, compounded=compounded)
- if period == "A" or any(x in period for x in ['year', 'eoy', 'yoy']):
+ if period == "A" or any(x in period for x in ["year", "eoy", "yoy"]):
return group_returns(returns, index.year, compounded=compounded)
- if 'week' in period:
+ if "week" in period:
return group_returns(returns, index.week, compounded=compounded)
- if 'eow' in period or period == "W":
- return group_returns(returns, [index.year, index.week],
- compounded=compounded)
+ if "eow" in period or period == "W":
+ return group_returns(returns, [index.year, index.week], compounded=compounded)
- if 'eom' in period or period == "M":
- return group_returns(returns, [index.year, index.month],
- compounded=compounded)
+ if "eom" in period or period == "M":
+ return group_returns(returns, [index.year, index.month], compounded=compounded)
- if 'eoq' in period or period == "Q":
- return group_returns(returns, [index.year, index.quarter],
- compounded=compounded)
+ if "eoq" in period or period == "Q":
+ return group_returns(
+ returns, [index.year, index.quarter], compounded=compounded
+ )
if not isinstance(period, str):
return group_returns(returns, period, compounded)
@@ -181,12 +177,12 @@ def to_excess_returns(returns, rf, nperiods=None):
if nperiods is not None:
# deannualize
- rf = _np.power(1 + rf, 1. / nperiods) - 1.
+ rf = _np.power(1 + rf, 1.0 / nperiods) - 1.0
return returns - rf
-def _prepare_prices(data, base=1.):
+def _prepare_prices(data, base=1.0):
"""Converts return data into prices + cleanup"""
data = data.copy()
if isinstance(data, _pd.DataFrame):
@@ -200,13 +196,12 @@ def _prepare_prices(data, base=1.):
data = to_prices(data, base)
if isinstance(data, (_pd.DataFrame, _pd.Series)):
- data = data.fillna(0).replace(
- [_np.inf, -_np.inf], float('NaN'))
+ data = data.fillna(0).replace([_np.inf, -_np.inf], float("NaN"))
return data
-def _prepare_returns(data, rf=0., nperiods=None):
+def _prepare_returns(data, rf=0.0, nperiods=None):
"""Converts price data into returns + cleanup"""
data = data.copy()
function = inspect.stack()[1][3]
@@ -218,15 +213,16 @@ def _prepare_returns(data, rf=0., nperiods=None):
data = data.pct_change()
# cleanup data
- data = data.replace([_np.inf, -_np.inf], float('NaN'))
+ data = data.replace([_np.inf, -_np.inf], float("NaN"))
if isinstance(data, (_pd.DataFrame, _pd.Series)):
- data = data.fillna(0).replace(
- [_np.inf, -_np.inf], float('NaN'))
- unnecessary_function_calls = ['_prepare_benchmark',
- 'cagr',
- 'gain_to_pain_ratio',
- 'rolling_volatility']
+ data = data.fillna(0).replace([_np.inf, -_np.inf], float("NaN"))
+ unnecessary_function_calls = [
+ "_prepare_benchmark",
+ "cagr",
+ "gain_to_pain_ratio",
+ "rolling_volatility",
+ ]
if function not in unnecessary_function_calls:
if rf > 0:
@@ -243,11 +239,10 @@ def download_returns(ticker, period="max", proxy=None):
params["start"] = period[0]
else:
params["period"] = period
- return _yf.download(**params)['Close'].pct_change()
+ return _yf.download(**params)["Close"].pct_change()
-def _prepare_benchmark(benchmark=None, period="max", rf=0.,
- prepare_returns=True):
+def _prepare_benchmark(benchmark=None, period="max", rf=0.0, prepare_returns=True):
"""
Fetch benchmark if ticker is provided, and pass through
_prepare_returns()
@@ -263,14 +258,17 @@ def _prepare_benchmark(benchmark=None, period="max", rf=0.,
elif isinstance(benchmark, _pd.DataFrame):
benchmark = benchmark[benchmark.columns[0]].copy()
- if isinstance(period, _pd.DatetimeIndex) \
- and set(period) != set(benchmark.index):
+ if isinstance(period, _pd.DatetimeIndex) and set(period) != set(benchmark.index):
# Adjust Benchmark to Strategy frequency
benchmark_prices = to_prices(benchmark, base=1)
- new_index = _pd.date_range(start=period[0], end=period[-1], freq='D')
- benchmark = benchmark_prices.reindex(new_index, method='bfill') \
- .reindex(period).pct_change().fillna(0)
+ new_index = _pd.date_range(start=period[0], end=period[-1], freq="D")
+ benchmark = (
+ benchmark_prices.reindex(new_index, method="bfill")
+ .reindex(period)
+ .pct_change()
+ .fillna(0)
+ )
benchmark = benchmark[benchmark.index.isin(period)]
benchmark.index = benchmark.index.tz_localize(None)
@@ -283,7 +281,7 @@ def _prepare_benchmark(benchmark=None, period="max", rf=0.,
def _round_to_closest(val, res, decimals=None):
"""Round to closest resolution"""
if decimals is None and "." in str(res):
- decimals = len(str(res).split('.')[1])
+ decimals = len(str(res).split(".")[1])
return round(round(val / res) * res, decimals)
@@ -296,12 +294,12 @@ def _in_notebook(matplotlib_inline=False):
"""Identify enviroment (notebook, terminal, etc)"""
try:
shell = get_ipython().__class__.__name__
- if shell == 'ZMQInteractiveShell':
+ if shell == "ZMQInteractiveShell":
# Jupyter notebook or qtconsole
if matplotlib_inline:
get_ipython().magic("matplotlib inline")
return True
- if shell == 'TerminalInteractiveShell':
+ if shell == "TerminalInteractiveShell":
# Terminal running IPython
return False
# Other type (?)
@@ -313,9 +311,9 @@ def _in_notebook(matplotlib_inline=False):
def _count_consecutive(data):
"""Counts consecutive data (like cumsum() with reset on zeroes)"""
+
def _count(data):
- return data * (data.groupby(
- (data != data.shift(1)).cumsum()).cumcount() + 1)
+ return data * (data.groupby((data != data.shift(1)).cumsum()).cumcount() + 1)
if isinstance(data, _pd.DataFrame):
for col in data.columns:
@@ -329,7 +327,9 @@ def _score_str(val):
return ("" if "-" in val else "+") + str(val)
-def make_index(ticker_weights, rebalance="1M", period="max", returns=None, match_dates=False):
+def make_index(
+ ticker_weights, rebalance="1M", period="max", returns=None, match_dates=False
+):
"""
Makes an index out of the given tickers and weights.
Optionally you can pass a dataframe with the returns.
@@ -365,7 +365,7 @@ def make_index(ticker_weights, rebalance="1M", period="max", returns=None, match
index = _pd.DataFrame(portfolio).dropna()
if match_dates:
- index=index[max(index.ne(0).idxmax()):]
+ index = index[max(index.ne(0).idxmax()) :]
# no rebalance?
if rebalance is None:
@@ -377,30 +377,30 @@ def make_index(ticker_weights, rebalance="1M", period="max", returns=None, match
# rebalance marker
rbdf = index.resample(rebalance).first()
- rbdf['break'] = rbdf.index.strftime('%s')
+ rbdf["break"] = rbdf.index.strftime("%s")
# index returns with rebalance markers
- index = _pd.concat([index, rbdf['break']], axis=1)
+ index = _pd.concat([index, rbdf["break"]], axis=1)
# mark first day day
- index['first_day'] = _pd.isna(index['break']) & ~_pd.isna(index['break'].shift(1))
- index.loc[index.index[0], 'first_day'] = True
+ index["first_day"] = _pd.isna(index["break"]) & ~_pd.isna(index["break"].shift(1))
+ index.loc[index.index[0], "first_day"] = True
# multiply first day of each rebalance period by the weight
for ticker, weight in ticker_weights.items():
index[ticker] = _np.where(
- index['first_day'], weight * index[ticker], index[ticker])
+ index["first_day"], weight * index[ticker], index[ticker]
+ )
# drop first marker
- index.drop(columns=['first_day'], inplace=True)
+ index.drop(columns=["first_day"], inplace=True)
# drop when all are NaN
index.dropna(how="all", inplace=True)
return index[index.index <= last_day].sum(axis=1)
-def make_portfolio(returns, start_balance=1e5,
- mode="comp", round_to=None):
+def make_portfolio(returns, start_balance=1e5, mode="comp", round_to=None):
"""Calculates compounded value of portfolio"""
returns = _prepare_returns(returns)
@@ -410,13 +410,13 @@ def make_portfolio(returns, start_balance=1e5,
p1 = to_prices(returns, start_balance)
else:
# fixed amount every day
- comp_rev = (start_balance + start_balance *
- returns.shift(1)).fillna(start_balance) * returns
+ comp_rev = (start_balance + start_balance * returns.shift(1)).fillna(
+ start_balance
+ ) * returns
p1 = start_balance + comp_rev.cumsum()
# add day before with starting balance
- p0 = _pd.Series(data=start_balance,
- index=p1.index + _pd.Timedelta(days=-1))[:1]
+ p0 = _pd.Series(data=start_balance, index=p1.index + _pd.Timedelta(days=-1))[:1]
portfolio = _pd.concat([p0, p1])