# st: RE: AW: xtmepoisson

 From Michael E Roettger To "statalist@hsphsun2.harvard.edu" Subject st: RE: AW: xtmepoisson Date Fri, 14 Aug 2009 11:40:39 -0400

```Have you examined the distribution of the dependent variable?

If xtmepoisson fails to converge while xtpoisson converges, you make have a distribution which does not fit the standard Poisson distribution.   Rabe-Hesketh and Skrondal (2008) note that xtpoisson adjusts for overdispersion due to the way the individual-level random effect is specified;  xtpoisson also assumes that the random effect has a log-gamma distribution.

If you have overdispersion, you might be able to use the xtgee command, specifying that the dependent variable is a negative binomial distribution.  Dummy variables may also be added to control for year effects.

Reference:
Rabe-Hesketh, S. and A. Skrondal. 2008. Multilevel and Longitudinal Modeling Using Stata. College Station, TX:  Stata Press.
________________________________________

-----Ursprüngliche Nachricht-----
Von: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] Im Auftrag von Simon Moore
Gesendet: Freitag, 14. August 2009 16:38
An: statalist@hsphsun2.harvard.edu
Betreff: st: xtmepoisson

Dear Statalist

I have data from over 60k school kids over ten years by school and I'm
interested in looking at the relationship between personal and aggregate
variables (alcprice, price of alcohol by year) on alcohol use (a count
variable, number of drinks). Data is collected from the same schools
over time.

I'm on a bit of a learning curve here and my initial thought was to use:

xtmepoisson alcohol age smoker alcprice ... || school: || year:
alcprice, cov(unstr)

where alcprice is the annual affordability of alcohol. Having started
the model three hours ago, it has now been sitting there at the
"Refining starting values:" without the slightest indication that it is

There is no such problem when I try:

xtpoisson alcohol age smoker alcprice year alcprice ... , i(school)

So my question is - have a done something strange in the xtmepoisson
command that Stata may object to, if not, is there a rule of thumb for
the length of time I should wait for?

Regards
Simon
*
*   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/

*
*   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/

*
*   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/
```