Jose,
If you're running multilevel models, you might also check out GLLAMM at:
www.gllamm.org
It's a free Stata module that allows you to estimate a bunch of
different models with random intercepts, slopes, or both. The
documentation and examples are pretty straightforward. The only
downside is that models with more than two levels or with random
slopes can take a while to run. You definitely want to make your data
set as small and tidy as possible before you hit "go."
Good luck,
Jay
On Sat, Feb 21, 2009 at 10:07 PM, JOSE A ALEMAN <[log in to unmask]> wrote:
> Dear list members,
>
> I would appreciate any help trying to implement Shor et. al's (2007) model
> for panel data in stata using a maximum likelihood or EM approach. Their
> model, as described in "A Bayesian Multilevel Modeling Approach to
> Time-Series Cross-Sectional Data," has random slopes for countries and
> years. I, however, am not sure if the proper stata syntax is
>
> xtmixed y x1 x2 || country || year
>
> or
>
> xtmixed y x1 x2 || country: || year:
>
> or something else.
>
> Thank you,
>
> Jose Aleman, Ph.D.
> Assistant Professor
> Political Science Department
> Fordham University
>
> **********************************************************
> Political Methodology E-Mail List
> Editors: Melanie Goodrich, <[log in to unmask]>
> Delia Bailey, <[log in to unmask]>
> **********************************************************
> Send messages to [log in to unmask]
> To join the list, cancel your subscription, or modify
> your subscription settings visit:
>
> http://polmeth.wustl.edu/polmeth.php
>
> **********************************************************
>
>
>
--
Jay Ulfelder, Ph.D.
Research Director
Political Instability Task Force
Science Applications International Corp. (SAIC)
[log in to unmask]
(301) 580-8736 [mobile]
**********************************************************
Political Methodology E-Mail List
Editors: Melanie Goodrich, <[log in to unmask]>
Delia Bailey, <[log in to unmask]>
**********************************************************
Send messages to [log in to unmask]
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
|