A consumer advocacy agency, Equitable Ernest, is interested in providing a service that allows an individual to estimate their own credit score (a continuous measure used by banks, insurance companies, and other businesses when granting loans, quoting premiums, and issuing credit). The file creditscore contains data from several thousand individuals. The variables in these data are listed in Problem 15.

Predict the individuals’ credit scores using a k-nearest neighbors. Set aside 20% of the data as a test set and use 80% of the data for training and validation.

a. Based on all the input variables, determine the value of k that minimizes the RMSE in a validation procedure.

b. Experiment with different subsets of variables as input features and re-calibrate the value of k to minimize the RMSE. How does this k-nearest neighbors model compare to the model obtained in part (a)?

c. For the best-performing k-nearest neighbors model in the validation procedure, what is the RMSE on the test set?