 |
| |||||
 |
| |||||||||||||||||||||
Fixed Effects Regression Models
Comment from the Stata technical groupFixed Effects Regression Models, by Paul D. Allison, is a useful handbook that concentrates on the application of fixed-effects methods for a variety of data situations, from linear regression to survival analysis. Fixed-effects models make less restrictive assumptions than their random-effects counterparts. For example, fixed-effects models allow unobservable variables to have whatever associations with the observed variables. As Allison points out, the individuals act as their own controls in a fixed-effects setting. That is not to say that fixed-effects models are not without their disadvantages. For example, the effects of covariates that are constant within individuals cannot be measured in this setting. However, such disadvantages should not dissuade one from using this powerful analysis technique that makes the most minimal of independence assumptions. In an appendix, Allison shows how to perform all the analyses by using Stata. Table of contents About the Author
Series Editor’s Introduction
1. Introduction
2. Linear Fixed Effects Models: Basics
The Two-Period Case
Extending the Difference Score Method for the Two-Period Case A First-Difference Method for Three or More Periods per Individual Dummy Variable Method for Two or More Periods per Individual Interactions With Time in the Fixed Effects Method Comparison with Random Effects Models A Hybrid Method Summary 3. Fixed Effects Logistic Models
The Two-Period Case
Three or More Periods Interactions With Time A Hybrid Method Methods for More Than Two Categories on the Response Variable Summary 4. Fixed Effects Models for Count Data
Poisson Models for Count Data With Two Periods per Individual
Poisson Models for Data With More Than Two Periods per Individual Fixed Effects Negative Binomial Models for Count Data A Hybrid Approach Summary 5. Fixed Effects Models for Events History Data
Cox Regression
Cox Regression With Fixed Effects Some Caveats The Hybrid Method for Cox Regression Fixed Effects Event History Methods for Nonrepeated Events Summary 6. Structural Equation Models With Fixed Effects
Random Effects as a Latent Variable Model
Fixed Effects as a Latent Variable Model A Compromise Between Fixed Effects and Random Effects Reciprocal Effects With Lagged Predictors Summary Appendix 1. Stata Programs for Examples in Chapters 2 to 5
Appendix 2. Mplus Programs for Examples in Chapter 6
References
Author Index
Subject Index
|
|||||||||||||||||||||
){kind=link}
){kind=link}