statisticalMethods.py

import pandas as pd
import requests
from sklearn.impute import SimpleImputer

def simpMovingAverage(df, n):

    df = df.transpose()

    df = df.rolling(n).mean()
    my_imputer=SimpleImputer()
    df = my_imputer.fit_transform(df)
    df = pd.DataFrame(df)


    return df

def expWeightFuncs(df, n):
    df = df.transpose()


    df = df.ewm(halflife = n).mean()
    my_imputer = SimpleImputer()
    df = my_imputer.fit_transform(df)
    df = pd.DataFrame(df)

    return df

def prevReturn(df):
    df.transpose()



def industryMovingAvg(df, n):
    hot_encoded = pd.get_dummies(df)
    instList = []
    epoch_res = requests.get('http://egchallenge.tech/instruments').json()

    for md in epoch_res:

        instList.append({
            'instrument_id': md['instrument_id'],
            'industry': md['industry'],
            'simpMovingAverage': simpMovingAverage(df, n)
        })

    instList.sort(key=lambda x: x[1])

    industry = instList[0][1]
    industryMovingSum = 0
    i=0

    totalIndustryAvgList = []

    for x in range(len(instList)):
        if instList[x][1] == industry:
            industryMovingSum += instList[x][2]
            i+=1

        else:
            totalIndustryAvgList.append({
                'industry': md['industry'],
                'totalIndustryAvg': industryMovingSum/i
            })

            industry = instList[x][i]
            i=1
            industryMovingSum += instList[x][2]

    return totalIndustryAvgList