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 on April 23, and its replacement, statalist.org is already up and running.


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

Re: st: xtnbreg Initial values


From   Ivan Png <iplpng@gmail.com>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: xtnbreg Initial values
Date   Wed, 12 May 2010 15:30:27 +0800

Many thanks, Rodolphe and Maarten.

The paper by Santos-Silva and Tenreyro explains when max likelihood
estimation of Poisson or similar specifications might fail.  Desbordes
below provides a way to identify the troublesome explanatory variables
and exclude them.

In my case, the solution was simpler.  Run OLS and get the
coefficients and use them as starting values in the xtnbreg

Specifically,

----

reg depvar indepvars
matrix _s=e(b)

xtnbreg depvar indepvars , fe vce(oim) iterate(20) from(_s, skip)


Once again, many thanks to STATA-listers!


On Fri, May 7, 2010 at 5:03 PM, Rodolphe Desbordes
<rodolphe.desbordes@strath.ac.uk> wrote:
> Ivan,
>
> Are you referring to this strategy?
>
> http://cep.lse.ac.uk/pubs/download/dp0932.pdf
>
>
> If that is correct, see http://privatewww.essex.ac.uk/~jmcss/panelpoisson.do
>
> Rodolphe
>
>
> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Ivan Png

> --
> Skype: ipng00
> T: +65 6516 6807
>
*
*   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