new exercise for regression setting

[hardin47]

Great example in Significance, June 2021. They talk about the regression line and the associated R^2 value with and without some of the outlying years. Would make a great problem. I think we could get the data, too.

One good thing about the outliers is that you can subset the years and talk about predictions only over a certain timeframe.

https://academic.oup.com/jrssig/article/18/3/32/7038195