STAT 440 - Forecasting
Lecture Review #14 - Ridge Regression

Review Questions:

What is meant by an unbiased estimator in statistics? Why do we like statistical procedures to be unbiased?

What is the basic idea behind ridge regression? That is, in practice how do we do ridge regression?

How do we decide what ridge constant ("c") should be used?

What are the advantages and disadvantages of using ridge regression?

Computational Exercises:

Use ridge regression to fit a model to the "Houses" data set.

a) What value of ridge constant ("c") did you use? Why?

b) What are the slopes of your ridge model, for the transformed data?

c) What is your model, transformed back to the original units?