Paper #613

Title:

Fixed and random effects in Classical and Bayesian regression

Author:

Silvio Rendón

Date:

April 2002

Abstract:

This paper proposes a common and tractable framework for analyzing different definitions of fixed and random effects in a contant-slope variable-intercept model. It is shown that, regardless of whether effects (i) are treated as parameters or as an error term, (ii) are estimated in different stages of a hierarchical model, or whether (iii) correlation between effects and regressors is allowed, when the same information on effects is introduced into all estimation methods, the resulting slope estimator is also the same across methods. If different methods produce different results, it is ultimately because different information is being used for each methods.

Keywords:

Bayes, panel data, nuisance parameters, fixed effects, random effects

JEL codes:

C11, C23

Area of Research:

Labour, Public, Development and Health Economics

Download the paper in PDF format