Refer to the scenario in Problem 42 regarding the identification of undecided voters. Apply a classification tree to classify observations as undecided or not by using Vote as the target (or response) variable. Set aside 20% of the data as a test set and use 80% of the data for training and validation.

a. Train a full classification tree and report its AUC from a validation experiment.

b. Train a pruned classification tree and report its AUC from a validation experiment.

c. Compare the AUC from the full classification tree in part

(a) to the AUC from the pruned classification tree in part (b). Explain the difference.

d. Consider a 50-year-old man who attends church, has 15 years of education, owns a home, is married, lives in a household of four people, and has an annual income of $150,000. For a cutoff value of 0.5, does the pruned tree classify this observation as Undecided?

e. For the pruned tree, compute and interpret the precision on the test set.

Problem 42

Campaign organizers for both the Republican and Democratic parties are interested in identifying individual undecided voters who would consider voting for their party in an upcoming election. A non-partisan group has collected data on a sample of voters with tracked variables. The variables in this data are listed in the following table.

Apply logistic regression with lasso regularization to classify observations as being undecided or not by using Vote as the target (or response) variable. Set aside 20% of the data as a test set and use 80% of the data for training and validation.

Transcribed Image Text:

Variable Vote Age HomeOwner Gender Married HouseholdSize Description Voter status: Undecided or decided Age of voter Binary variable indicating whether voter owns home Gender identity provided by voter Marital status of voter Number of people in voter's household Annual income (in $1000) Income Education Years of education Religious Binary variable indicating whether voter regularly attends religious services