Heckman two-step postestimation

[Max Resende]

Dear Stata users,

I fitted my model using two-step estimator by -Heckman and I need help with the postestimation comands:

1).I I want to calculate the average marginal effects for the regression equation, but when i type:

 margins, dydx(*) predict(ycond)

i got the following error message:

predict option ycond not appropriate with margins

The strange thing is that I was able to estimate the marginal effects for the selection equation using

margins, dydx(*) predict(psel)

it worked just fine.

2) After I calculate the marginal effects for both equation, any suggestion on graphics instead:

marginsplot

If anyone could help me I appreciate it.

Max

[Phil Bromiley]

You'll increase your chances of a useful answer by following the FAQ on asking questions - include Stata code in code delimiters, readable Stata output, and sample data using dataex.

The Heckman Postestimation entry in the manual says, "ycond and yexpected are not allowed with margins after heckman, twostep." So, you can use the ml version of heckman, or you can do dydx based on the linear prediction. You could use predict to see if predict with ycond is much different than xb.

[KIM SUJIN]

I also wonder how to get