Re: st: ARMA(1,1) with Multiple Panels

[Alexander Gelber <gelber@fas.harvard.edu>]

Many thanks for this extremely helpful suggestion!

When you say that the "conditional estimator is conditional on simple
rules for starting the process at the beginning of the sample," what
simple rules or assumptions, in particular, are you referring to?  I have
had trouble finding the answer using Stata help.

Best,

Alex

On Tue, 22 Aug 2006, Vince Wiggins, StataCorp wrote:

> Alexander Gelber <gelber@fas.harvard.edu> asks,
>
> > Which Stata commands should I use to estimate an ARMA(1,1) process
> > in a panel dataset with multiple panels?
> >
> > The problem is that when I just use the normal command to estimate
> > an ARMA(1,1), it gives me an error message:
> >
> >  arima dep ind, arima(1,0,1)
> > sample may not include multiple panels
> > r(459);
>
> Alexander appears willing to assume that the same process holds across his
> panels.  In that case, he can add the -condition- option to his -arima-
> command to produce conditional rather unconditional maximum likelihood
> estimates of the parameters.
>
> These are simply different estimators of the same parameters, but the
> conditional estimator is conditional on simple rules for starting the process
> at the beginning of the sample, whereas the unconditional estimator starts the
> process using the optimal solution from the current parameter estimates.  This
> means that the conditional estimator is amenable to restarting the time-series
> process at each panel break whereas the unconditional estimator is not.
>
>
> -- Vince
>    vwiggins@stata.com
>
> *
> *   For searches and help try:
> *   http://www.stata.com/support/faqs/res/findit.html
> *   http://www.stata.com/support/statalist/faq
> *   http://www.ats.ucla.edu/stat/stata/
>
*
*   For searches and help try:
*   http://www.stata.com/support/faqs/res/findit.html
*   http://www.stata.com/support/statalist/faq
*   http://www.ats.ucla.edu/stat/stata/