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: xtivlogit, fe or ivclogit - Why not?


From   Austin Nichols <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: xtivlogit, fe or ivclogit - Why not?
Date   Fri, 3 Dec 2010 10:46:39 -0500

There is not even an ivlogit; see e.g.
http://www.econstor.eu/dspace/bitstream/10419/23743/1/4-05.pdf
Angrist would have you run linear IV; see e.g.
http://www.mostlyharmlesseconometrics.com
Wooldridge proposes a variety of approaches which respect the
nonlinearity: start with
http://www.stata.com/meeting/snasug08/abstracts.html#wooldridge
But you could also run -ivprobit- with dummies, for another biased
estimator to compare to; see also
http://www.stata.com/statalist/archive/2009-08/msg01593.html

On Fri, Dec 3, 2010 at 10:08 AM,  <C.Peterson1@lse.ac.uk> wrote:
> Dear StataList,
>
> 1) Would anyone care to offer a brief explanation of why an instrumental
> variables estimate cannot be made for fixed effects panel data with a
> binary outcome? Stata has no xtivlogit, fe or ivclogit options.
>
> 2) Do you have any suggestions about how to deal with potential
> endogeneity of a continous independent variable in model of panel data
> with a binary outcome, in which the Hausmann test indicates fixed
> effects estimation?
>
*
*   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