You can load the data in R, #load the data

library(MASS)

data("Boston")

Problem 4. The Boston dataset in the MASS package describes housing information for 506 different suburbs of Boston. We are interested in using these data to predict whether or not a new suburb is above or below the median per capita crime rate. To view the descriptions of the varibles, run ?Boston after loading the dataset to open the docu- mentation page that will appear in the Help pane. For example, the documentation page indicates that the variable crim is the per capita crime rate by town. 3SR Dalal, Fowlkes EB, and Hoadley B. Risk analysis of the Space Shuttle: Pre-Challenger Prediction of Failure. Journal of the American Statistical Association. 84: 408 (1989). Using a validation approach, investigate which of the following models is the best predictive model: - Model 1: a single-predictor model with medv Model 2: all variables except zn, indus, and nox Model 3: all variables except Istat, rm, and chas - Model 4: all variables Model 5: all variables, including an interaction term between chas and 1stat and an inter- action term between chas and rad Summarize your approach and findings