xtfevd

Thomas Plümper and Vera Troeger

Efficient Estimation of Time-Invariant and Rarely Changing Variables
in Finite Sample Panel Analyses with Unit Fixed Effects
 

Political Analysis Vol. 15:2, 124-139.
 

This paper suggests a three-stage procedure for the estimation of time-invariant and rarely  changing variables in panel data models with unit effects. The first stage of the proposed  estimator runs a fixed-effects model to obtain the unit effects, the second stage breaks  down the unit effects into a part explained by the time-invariant and/or rarely changing variables and an error term, and the third stage re-estimates the first stage by pooled OLS (with or without autocorrelation correction and with or without panel-corrected SEs) including  the time-invariant variables plus the error term of stage 2, which then accounts for the  unexplained part of the unit effects. We use Monte Carlo simulations to compare the finite sample properties of our
estimator to the finite sample properties of competing estimators. In doing so, we demonstrate that our proposed technique provides the most reliable estimates under a wide variety of specifications common to real world data.

Download the beta version of the ado and the help-file (for Stata 7.0-9.x)

xtfevd.zip

You may also download unzipped versions of the ado und the hlp file seperately

xtfevd.ado
xtfevd.hlp

Save ado and hlp files into a
Stata/ado/x folder
and type help xtfevd onto the Stata commend window to get an overview of the command structurre and  the options.

Last changes, September 26, 2007
(please do not use earlier versions of the estimator).

17 May 07: Bug detected! If the order of the invariant variables in the main command and in the invariant bracket is not identical, the estimator assigns incorrect standard errors to the invariant variables. We will correct the bug asap.

21 May 07: Problem fixed  (well, kind of ...). The new version of the ado (May07) restrict the organization of variables so that the reported standard errors are correct. We admit that this solution is inconvenient and work on a more convenient one, but that will take a while.

26 September 07: Problem finally fixed. You may use the new version of the estimator (Version 2.00 beta; September 26, 2007) without paying attention to the order of the variables.