One rarely, if ever, sees ARIMA models being estimated with panel data. I believe the reason is that in the great majority of instances, one has data for a small number of waves, and consequently does not have enough time points to estimate such models with an acceptable degree of precision, especially those with monthly components. David Greenberg, Sociology Department, New York University
----- Original Message -----
From: Jimmy Verner <[email protected]>
Date: Tuesday, October 7, 2008 2:07 pm
Subject: st: ARIMA for panel data?
To: [email protected]
> I'm working with cross-sectional time-series data. X-12 ARIMA has a
>
> feature that allows the software to pick the optimal model for any
> particular series, then apply that model to all series. However, the
>
> software cannot pick the optimum model for a collection of series. So
>
> far as I am aware, Stata does not have such a feature, either. Is
> that correct? Is there any software that would suggest a single model
>
> (i.e, same form with identical coefficients) for all series rather
> than only one series?
>
> TIA
>
> Jimmy Verner
> Graduate Student
> University of Texas - Dallas
>
> *
> * 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/
