[Mustafa Coban]

Hi,

a quick note for those who are interested in treatment effects from recursive bivariate probit models. The package rbiprobit provides the computation of 3 different treatment effects for recursive bivariate probit models

ssc install rbiprobit

Quick example for ATE, ATET and ATEC

rbiprobit y1 = x1 x2 x3, endog(y2 = x2 x3 x4)
rbiprobit tmeffects, tmeff(ate)
rbiprobit tmeffects, tmeff(atet)
rbiprobit tmeffects, tmeff(atec)

[Mustafa Coban]

Hi James,

I agree that the ATE in a recrusive bivariate probit model can be estimated using this workaround, but the calculation of the ATET is incorrect following Greene (Econonmetric Analysis, 2018, p. 819).

Since you want to fit a recursive bivariate probit model, I would recommend to switch to the rbiprobit package; available on SSC by

ssc install rbiprobit

This package accounts for recursive nature of your model, gives correct predictions and enables the estimation of various treatment effects. Here, a short example for the estmation of the ATE and ATET

rbiprobit y1 = x1 x2 x3, endog(y2 = x1 x2 x4 x5)
rbiprobit tmeffects, tmeffect(ate)
rbiprobit tmeffects, tmeffect(atet)

For more background:

Check out the repo of the package on GitHub and the current presentation slides from the Italian Stata Conference.