femlogit—implementation of the multinomial logit model with fixed effects

Abstract

Fixed-effects models have become increasingly popular in social-science research. The possibility to control for unobserved heterogeneity makes these models a prime tool for causal analysis. Fixed-effects models have been derived and implemented for many statistical software packages for continuous, dichotomous, and count-data dependent variables. Chamberlain (1980, Review of Economic Studies 47: 225–238) derived the multinomial logistic regression with fixed effects. However, this model has not yet been implemented in any statistical software package. Possible applications would be analyses of effects on employment status, with special consideration of part-time or irregular employment, and analyses of effects on voting behavior that implicitly control for long-time party identification rather than measuring it directly. This article introduces an implementation of this model with the new command femlogit. I show its application with British election panel data.