Tobit models are used to model variables subject to exogenous censoring. For example, duration data cannot be observed longer than the survey period; hours of work cannot be observed negative although an individual might be better off consuming more leisure time than is available. This article reviews a list of econometric techniques to estimate Tobit models. Maximum likelihood, Heckman’s two-stage estimator, and Powell’s trimmed least squares are successively addressed.
KeywordsCensored regression model Endogenous regressors Heteroskedasticity Labour supply Maximum likelihood Non-normal errors Ordinary least squares Probit model Tobit model Trimmed least squares Two-stage estimation
JEL ClassificationsC25 C24
- Pudney, S. 1989. Modelling individual choice: The econometrics of corners, kinks and holes. Oxford: Blackwell.Google Scholar