CD Real Estate specializes in residential real estate services in the state of California. To complement the experience and local market knowledge of its licensed realtors, CD Real Estate wants to develop an analytical tool to predict the value of real estate. The file calireal contains data on some census tracts in California. The variables in these data are listed in the following table.

Predict the median house value using an individual regression tree. Set aside 50% of the data as a test set and use 50% of the data for training and validation.

a. Train a full regression tree and report its RMSE from a validation experiment.

b. Train a pruned regression tree and report its RMSE from a validation experiment.

c. Compare the RMSE from the full regression tree in part (a) to the RMSE from the pruned regression tree in part (b). Explain the difference.

d. Compute the RMSE of the pruned regression tree from part (b) on the test set.

e. Consider a tract with 5 credit bureau inquiries, that has a Longitude = 2117.9, Longitude = 33.64, Age = 36, Rooms = 2,107, Bedrooms = 357, Population = 850, Households = 348, and Income = 5.0532. Using the pruned tree settings from part (b), what is the predicted median house value for this tract?

Transcribed Image Text:

Variable Longitude Description Longitude in center of tract Latitude Age Latitude in center of tract Median age of housing Rooms Total number of rooms Bedrooms Total number of bedrooms Population Population of tract Households Number of households Income Value Median income Median house value of tract (in $1,000s)