Bookmark and Share

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: Re: IVPOIS: Endogeneity Test


From   Joseph Monte <[email protected]>
To   [email protected]
Subject   Re: st: Re: IVPOIS: Endogeneity Test
Date   Tue, 3 Jul 2012 08:28:42 +0100

Hi Statalisters,

Austin,is it possible for you to build the exogeneity test into
-ivpois- as you promised to do below?

Is there a negative binomial equivalent of -ivpois- since I believe
-ivpois- pertains to a poisson regression while I am interested in
running a negative binomial regression in the instrumental variable
framework? Hence, I am referring to something on the lines of
-ivnbreg- (which does not exist at the moment). I have a case of
over-dispersion (mean of count variable = 79, variance of same count
variable = 2003), hence I believe that -ivpois- is not appropriate.

Thanks,

Joe



On Mon, Jan 14, 2008 at 3:24 PM, Austin Nichols <[email protected]> wrote:
> There is no such test built into -ivpois- (on SSC) though it would be
> trivial to add; I'll put it on my list.  An easy way to test by hand
> is to predict a residual in the first stage and test its significance
> in the second stage (see e.g. Wooldridge
> http://stata.com/bookstore/cspd.html pp.664-665).
>
> use http://fmwww.bc.edu/ec-p/data/wooldridge/card, clear
> loc x "exper* smsa* south mar black reg662-reg669"
> * ivpois wage `x', endog(educ) exog(nearc4)
> * test of exogeneity; under the null, test is correct
> reg educ `x' nearc4
> predict vhat, resid
> poisson wage `x' educ vhat
> test vhat
>
> BTW, I don't think you posted this successfully to Statalist, though
> the email is copied to [email protected]; are you
> subscribed?  Relying on any one specific Statalister to answer a
> question is a risky strategy at best....
>
> On Jan 12, 2008 11:30 AM, ippab <[email protected]> wrote:
>> Hi Austin,
>>
>> Is there anyway to test for endogeneity in IVPOIS?  How do we know whether
>> we should use the estimates from the original model (Poisson model) or from
>> IVPOIS?  I guess I am thinking of something like Hausman test or Wald test
>> of exogeneity.
>>
> *
> *   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/help.cgi?search
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index