Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: IVPOIS -- Error: Hessian is not positive semidefinite.


From   "Austin Nichols" <austinnichols@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   st: Re: IVPOIS -- Error: Hessian is not positive semidefinite.
Date   Tue, 18 Dec 2007 17:43:20 -0500

Mohammad Rahman <ippab.research@gmail.com>:
It should be able to run on a large dataset, but the SEs are computed
via bootstrap, so it takes a while.  You might see how it scales up by
estimating on a random 10th of the data, then half the data, and
specify the option
 reps(5)
to just do 5 -bootstrap- replicates (the default is 50).  Someday soon
I will add analytic SE calculations for the impatient.

On Dec 18, 2007 5:36 PM, Mohammad Rahman <ippab.research@gmail.com> wrote:
> Thanks so much for your quick reply.  I have used the following commands
> from your email on 12 Nov 2007:
> ssc inst ivpois
> sysuse auto, clear
> g manuf=word(make,1)
> ivpois mpg wei, exog(turn) endog(disp) cl(manuf)
>
> However, right after emailing you, I think I found the solution to the
> problem.  Basically, one needs to update Stata 10 before one can
> successfully use IVPOIS.
>
> Can IVPOIS be used for a large dataset (sample size: 165,000)?  I have been
> running it for a while and it is still not doing anything.  It is stuck at
> "running iv_poisson on estimation sample".
*
*   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/



© Copyright 1996–2021 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index