Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down at the end of May, and its replacement, statalist.org is already up and running.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: st: ivpois with a binary endogenous predictor


From   Austin Nichols <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: ivpois with a binary endogenous predictor
Date   Thu, 20 Sep 2012 22:00:54 -0400

Dimitriy V. Masterov <dvmaster@gmail.com>:
No need to do anything different, though you can potentially improve
efficiency by running a logit of x on z first, predict p, then use p
as your excluded instrument. Instead of -ivpois-, you may prefer
-gmm-, assuming you are using Stata >10.
See also e.g.
http://www.stata.com/statalist/archive/2009-07/msg01256.html
on comparing your -gmm- estimates to others.

On Thu, Sep 20, 2012 at 9:31 PM, Dimitriy V. Masterov
<dvmaster@gmail.com> wrote:
> I would like to estimate the effect of an endogenous binary variable
> (x) on expenditure and/or transactions (y), which have a fair amount
> of zeros (~12%) and a long right tail. I have a continuous instrument
> (z). I was hoping to use the  Austin Nichols' -ivpois- command to do
> this.
>
> Are there any econometric issues with the first stage being very
> different beast from the outcome equation?
>
> How does this issue interact with doing the diagnostics by hand, like
> the Hausman endogeniety test using the residuals?
>
> Are there better ways to tackle this problem?
*
*   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/


© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   Site index