IZA

Logo
Identification and Decompositions in Probit and Logit Models
by Chung Choe, Seeun Jung, Ronald L. Oaxaca
(January 2017)

Abstract:
Probit and logit models typically require a normalization on the error variance for model identification. This paper shows that in the context of sample mean probability decompositions, error variance normalizations preclude estimation of the effects of group differences in the latent variable model parameters. An empirical example is provided for a model in which the error variances are identified. This identification allows the effects of group differences in the latent variable model parameters to be estimated.
Text: See Discussion Paper No. 10530