A fixed effects model would be appropriate, except that you lose the
ability to learn about the effects of any independent variables that do
not change over time. David Greenberg
----- Original Message -----
From: joel miller <[email protected]>
Date: Monday, February 27, 2006 4:43 pm
Subject: Re: st: fixed effects on PCSE model
> Thanks David, that�s good to know.
>
> But, do you know what choice of models I do have when dealing with
> a small T
> and an N that is much larger than T? I am keen, from a theoretical
> point of
> view, to use a fixed effects approach, and I had read that Beck
> Katz model
> would be better for my situation than the Park model. Should I just
> be
> thinking OLS fixed effects?
>
> Thanks,
>
> Joel.
>
>
> >From: David Greenberg <[email protected]>
> >Reply-To: [email protected]
> >To: [email protected]
> >Subject: Re: st: fixed effects on PCSE model
> >Date: Mon, 27 Feb 2006 16:34:16 -0500
> >
> >Panel-corrected standard error estimation is called for when the
> number>of time points is large, and the number of cases is small.
> That is not
> >your circumstance, and themethod of analysis you have chosen is
> >inappropriate. David Greenberg, Sociology Department, New York
> University.>
> >----- Original Message -----
> >From: joel miller <[email protected]>
> >Date: Monday, February 27, 2006 10:14 am
> >Subject: st: fixed effects on PCSE model
> >
> > > I�m doing some TSCS modeling on police activity data that has 42
> > > separate
> > > police areas, and about 6 or 7 time points. I�m thinking that PSCE
> > > models
> > > are most appropriate here (though I�d welcome any comments or
> > > detractions
> > > from this). I�m also keen to try these models with unit fixed
> effects.> >
> > > My key question is, does Stata have a default commands to create
> > > and include
> > > dummy variables in the .xtpcse commands, or will I have to create
> > > my own
> > > dummies. In other Stata PCSE I know how to build in fixed effects,
> > > but I
> > > haven�t found syntax for this with the .xtpcse modelling
> procedure.> >
> > > Thanks,
> > >
> > > Professor Joel Miller
> > > University of M�laga, Spain
> > >
> > >
> > > *
> > > * 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/
>
>
> *
> * 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/
