hecman模型

The Heckman selection model is a two-equation model.

 

There are two popular estimation methods for the model: maximum likelihood (full-information maximum likelihood, FIML) and the two-step procedure (limited-information maximum likelihood, LIML) (Leung and Yu, 1996)[33。 the LIML approach, is preferred and widely used as an alternative. the estimation of the Heckman selection regression starts from the selection model. In the first step, the probit regression is used to model the sample selection process in Eq 5, and then the inverse Mills ratio λ, the error from the probit equation explaining selection, is calculated based on the probit regression results. In the second step, the inverse Mills ratio is added to multiple regression analysis as an independent variable, and ordinary least square is used to provide the consistent parameter estimates in Eq 4.