Stata 5: How can I get the Mills' ratios for my Heckman model?

The heckman command does not save the Mills’ ratios that it uses in the estimation of the Heckman selection model. However, it is straightforward to calculate them yourself:

1. Calculate the probit model for the participation.
2. Obtain the linear predictors from the model. For example,

    . predict phat, xb
   

3. Calculate the Mills' ratios. For example,

    . gen mills = exp(-.5*phat2)/(sqrt(2*_pi)*normprob(phat))
   

Alternatively, you could specify the trace option to heckman.