Take the Boston Data Set and make 2-3 variations on the models so far presented. See if you can find one that improves over the ones presented to date. For example, perhaps including more variables will help. Other things you can do For some variables, change s to bs (or vice-versa) For a bs variable, change df to increase or decrease the number of knot points Change a splined variable( s or bs) to a linear response (or vice-versa). Revisit your linear model from part e of exercise 1 in homework 1. Make a GAM you like better. Load the data set diamonds from the ggplot2 library. Using only the numerical features, fit a linear regression model to predict the price. Using visualization, determine which is any of the factors you wish to recode as numerics. If so, make that recode (you should be able to figure this out, but you can always google it or ask me). With the recoded variables, find a linear model you like to predict price