Chapter 28. Automatic Variable Selection

Table of Contents

Introduction
Method description
Method assumptions
Full Model
Forward Selection
Backward Elimination
Stepwise Selection
Best Subset Selection
Usage
Data requirements
Model building and testing
Model application
Example of automatic variable selection
References

Introduction

This chapter describes automatic variable selection methods available in the Linear Regression, Logistic Regression, Survival Analysis, Time Series and Bivariate Probit modules. These methods are intended to help the user built a well-fitted final model by efficiently analysing numerous alternative, intermediate models.