Sean,
-yexpected- is not an option for predict following -treatreg-
Perhaps, what you want is -yctrt-, which is the E(y | treatment = 1)
or -ycntrt-, E(y | treatment = 0).
Scott
----- Original Message -----
From: "Sean Michael Bolks" <[email protected]>
To: <[email protected]>
Sent: Friday, January 02, 2004 8:45 AM
Subject: st: marginal effects for treatment models
> I have two questions for the list. I have been using the treatreg command
> to examine a treatment effects model. I am interested in obtaining the
> marginal effecets of the independent variables of all equations for the
> expected value of the dependent variable. I realize that the marginal
> effect of a variable is composed of the selection equation and outcome
> equation. For a Heckman model, the marginal effect transformation is
>
> dE[y| z*>0]/dx = beta - (alpha*rho*simga*delta(alpha))
>
> where delta(alpha) = inverse Mills' ratio *(inverse Mills' ratio *
> selection prediction)
>
> The overall treatment effect employs a similar transformation using the
> condition of the treatment minus the condition of no treatment.
>
> However, how does one calculate the marginal effects for particular
> variables in a treatment model?
>
> Second, I have looked at the mfx command. There is the following for
> heckman:
>
> mfx compute, predict(yexpected)
>
> This would give me what I am looking for, but it does not appear to be
> an option for treatreg--I get the following message:
>
> option yexpected not allowed
>
> Any suggestions would be greatly appreciated.
>
> Sean Bolks
>
>
> *
> * For searches and help try:
> * http://www.stata.com/support/faqs/res/findit.html
> * http://www.stata.com/support/statalist/faq
> * http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
