Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: st: Treatreg with probit
From
Austin Nichols <[email protected]>
To
[email protected]
Subject
Re: st: Treatreg with probit
Date
Tue, 5 Jul 2011 17:42:28 -0400
John Antonakis <[email protected]>:
You have a binary outcome and a single binary endogenous variable?
With excluded instruments?
This is sometimes called the triangular simultaneous equation model,
and -biprobit- handles it, as does -cmp-.
Do not manually include a hazard estimate in a second stage probit!
N.B. Heckman-type models with some missing outcomes are essentially
equivalent to treatment-effect-type models;
compare methods and formulas in e.g. [R] treatreg and [R] heckman,
esp. for two-step estimators--
one augments the regression with the first-stage hazard (generalized
residual) in either case.
On Tue, Jul 5, 2011 at 5:18 PM, John Antonakis <[email protected]> wrote:
> Hi:
>
> Yip; I know of -cmp- and have recommended it in the past to others. It does
> do Heckman-type tobit models (with censoring on y), though it doesn't seem
> to me that it can do treatment effect-type models (where y is observed in
> the treated and non-treated group)--or did I misread the help file?
>
> Best,
> J.
>
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/