Missing data handling

[Johannes_Auer]

I am also in the process of developing a Heckman model. It is running, but the coefficients of the selection equation are still off.
Maybe it helps you as a starting point for your model:

Heckman sample selection model (working but biased selection coefficients)

Dear Stan-users, I am in the process of developing a simple heckman selection model (aka Tobit Type 2 model) using the Wooldridge data set “mroz” (Wooldridge, 2002, page 565, example 17.6). The data set contains the wages for married females and a set of covariates. However, we only observe wages for 428 women who choose to work. We do not observe the wages for the 325 women who are not working. The heckman model works with a probit selection equation to capture whether or not a woman is worki…