Linear Regression Models for Panel Data Using SAS, STATA, LIMDEP, and SPSS

Table of Contents

• Introduction
• 1.1 Data Arrangement
• 1.2 Fixed Effect versus Random Effect Models
• 1.3 Estimation and Software Issues
• Least Squares Dummy Variable Regression
  • 2.1 Model 1 without a Dummy Variable
  • 2.2 Model 2 with a Dummy Varialbe
  • 2.3 Visualization of Model 1 and 2
  • 2.4 Alternatives to LSDV1
  • 2.5 Estimating Three LSDVs
• Panel Data Models
  • 3.1 Functional Forms and Notation
  • 3.2 Fixed Effect Models
  • 3.3 Random Effect Models
  • 3.4 Hausman Test: Fixed Effects versus Random Effects
  • 3.5 Poolability Test
• The Fixed Group Effect Model
  • 4.1 The Pooled OL Regression Model
  • 4.2 LSDV1 without a Dummy
  • 4.3 LSDV2 without the Intercept
  • 4.4 LSDV3 with Restrictions
  • 4.5 Within Group Effect Model
  • 4.6 Between Group Effect Model: Group Mean Regression
  • 4.7 Testing Fixed Group Effects
  • 4.8 Summary
• The Fixed Time Effect Model
  • 5.1 Least Squares Dummy Variable Models
  • 5.2 Within Time Effect Model
  • 5.3 Between Time Effect Model
  • 5.4 Testing Fixed Time Effects
• The Fixed Group and Time Effect Model
  • 6.1 Least Squares Dummy Variable Models
  • 6.2 LSDV1 without Two Dummies
  • 6.3 LSDV1 + LSDV3
  • 6.4 LSDV3 with Two Restrictions
  • 6.5 Two-way Within Effect Model
  • 6.6 Using the TSCSREG and PANEL Procedures
  • 6.7 Testing Fixed Group and Time Effects
• Random Effect Models
  • 7.1 The One-way Random Group Effect Model
  • 7.2 Estimations in SAS, STATA, and LIMDEP
  • 7.3 The One-way Random Time Effect Model
  • 7.4 The Two-way Random Effect Model in SAS
  • 7.5 Testing Random Effect Models
  • 7.6 Fixed Effects versus Random Effects
  • 7.7 Summary
• Poolability Test
• Conclusion
• Appendix
• References


This document summarizes linear regression models for panel data and illustrates how to estimate each model using SAS 9.1, STATA 9.0, LIMDEP 8.0, and SPSS 13.0. This document does not address nonlinear models (i.e., logit and probit models), but focuses on linear regression models.