SAS Statistics by Example

Chapter 9  Multiple Regression

Introduction

Fitting Multiple Regression Models

Running All Possible Regressions with n variables

Producing Separate Plots Instead of a Panel

Choosing the Best Model (Cp and Hocking’s Criteria)

Forward, Backward, and Stepwise Selection Methods

Forcing Selected Variables into a Model

Creating Dummy (Design) Variables for Regression

Detecting Collinearity

Influential Observations in Multiple Regression Models

Conclusions

Introduction

This chapter covers multiple regression models. You will learn how to generate diagnostics to help select variables for a model as well as how to perform stepwise techniques. The same diagnostics that are available with simple linear regression can be used with multiple regression models. ...