Predicting Housing Median Prices. The file BostonHousing.xls contains information on over 500 census tracts in Boston, where for each tract 14 variables are recorded. The last column (CAT.MEDV) was derived from MEDV, such that it obtains the value 1 if MEDV>30 and 0 otherwise. Consider the goal of predicting the median value (MEDV) of a tract, given the information in the first 13 columns.

Partition the data into training (60%) and validation (40%) sets.

a. Perform a k-NN prediction with all 13 predictors (ignore the CAT.MEDV column), trying values of k from 1 to 5. Make sure to normalize the data (click “normalize input data”). What is the best k chosen? What does it mean?

b. Predict the MEDV for a tract with the following information, using the best k:

(Copy this table with the column names to a new worksheet and then in “Score new data” choose “from worksheet.”)

c. Why is the error of the training data zero? d. Why is the validation data error overly optimistic compared to the error rate when

applying this k-NN predictor to new data?

e. If the purpose is to predict MEDV for several thousands of new tracts, what would be the disadvantage of using k-NN prediction? List the operations that the algorithm goes through in order to produce each prediction.

(Shmueli 146-147)

Transcribed Image Text:

CRIM ZN 0.2 TAX 307 O PTRATIO 21 INDUS 7 B 360 CHAS O LSTAT 10 NOX 0.538 RM 6 AGE 62 DIS 4.7 RAD 4 CRIM ZN 0.2 TAX 307 O PTRATIO 21 INDUS 7 B 360 CHAS O LSTAT 10 NOX 0.538 RM 6 AGE 62 DIS 4.7 RAD 4

Answer rating: 100% (QA)

KNearest Neighbor prediction method In this method the Training Set is used to predict the value of ... View the full answer