Statalist


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

Re: st: mvprobit and changing results


From   "Stas Kolenikov" <skolenik@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: mvprobit and changing results
Date   Mon, 21 Jul 2008 17:33:56 -0500

-mvprobit- is a simulation-based estimator. You need to fix the random
number generator seed, see the corresponding option of -mvprobit-.
However if you see drastic differences between runs, that probably
means that the simulation variability dominates the underlying
sampling variability, so you would want to increase the number of
-draws-. I remember having a discussion with Stephen Jenkins about the
number 5 which I found ridiculously low, but he in the end convinced
me that it works OK in most applications. Yours might be an exception
-- increase that number until the results with different -seed- values
become more stable.

On Mon, Jul 21, 2008 at 12:30 PM, U.G. Narloch <ugn20@cam.ac.uk> wrote:
> Dear Statalist Users,
>
> I use two do-files: the first one for generating the dataset from different
> dta-files and the second one for running the data analysis focusing on a
> mv-probit model drawing on the dataset generated from the first do-file.
>
> I have just realised that every time I replace the previous dataset.file by
> simply rerunning the first do-file the mv-probit results for exactly the
> same model specifiaction change. Coefficients and standard errors vary
> slightly, but the output for the correlation coefficients between error
> terms change significantly (sometimes even from negative to positive
> significance).
>
> As the output for other estimation commands (reg, probit, etc.) does not
> change I guess that this is due to the mvprobit. But I do not understand
> what could be the reason for this.
>
> Is there anyone having an idea what is going on?
>
> I am grateful for any comments!
>
> Thanks,
> Ulf Narloch
>
>
>
>
> *
> *   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/
>



-- 
Stas Kolenikov, also found at http://stas.kolenikov.name

Small print: Please do not reply to my Gmail address as I don't check
it regularly.
*
*   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   |   What's new   |   Site index