Home  >>  Archives  >>  Volume 6 Number 4  >>  st0112

The Stata Journal
Volume 6 Number 4: pp. 461-481



Subscribe to the Stata Journal
cover

Practical fixed–effects estimation methods for the three–way error–components model

Martyn Andrews
School of Economic Studies
University of Manchester
Manchester, UK
martyn.andrews@man.ac.uk
Thorsten Schank
Department of Economics
Friedrich-Alexander-Universität Erlangen–Nürnberg
Nürnberg, Germany
thorsten.schank@wiso.uni-erlangen.de
Richard Upward
School of Economics
University of Nottingham
Nottingham, UK
richard.upward@nottingham.ac.uk
Abstract.   Methods for fixed-effects estimation of the three-way error-components model are not yet standard. Where possible, we make the fixed-effects methods originally developed by Abowd, Kramarz, and Margolis (Econometrica 67: 251–333) for linked worker–firm data more accessible. We also show how these methods can be implemented in Stata. There is a caveat, however. If the researcher wants to recover estimates of the error components themselves, and the number of units at the higher level of aggregation is large, memory or matrix constraints may make using Stata to estimate the components themselves infeasible.
Terms of use     View this article (PDF)

View all articles by these authors: Martyn Andrews, Thorsten Schank, Richard Upward

View all articles with these keywords: grouping, linked employer–employee panel data, fixed effects

Download citation: BibTeX  RIS

Download citation and abstract: BibTeX  RIS