In addition to Doug's advice, I would suggest you take a look at the September
5th Statalist message by Bill Gould, Vince Wiggins, and David Drukker at:
http://www.stata.com/statalist/archive/2003-09/msg00103.html
Scott
----- Original Message -----
From: "Doug Mahony" <[email protected]>
To: <[email protected]>
Sent: Tuesday, January 06, 2004 2:46 PM
Subject: Re: st: panel data with binary dependent variable
>
> Hi Grazia,
>
> Look at the xt family of commands for time series panel data. In particular
xtlogit. For example the command for your regression would look like this:
xtlogit y x1 x2 x3, i(firm) fe
>
>
>
> You will also find a host of diagnostics in the manuals or in the help file.
>
> Hope this helps,
>
> Doug
>
>
>
>
> ---------- Original Message ----------------------------------
> From: [email protected]
> Reply-To: [email protected]
> Date: Tue, 6 Jan 2004 20:17:45 +0100
>
> I have an unbalanced panel of firms and times and three other explanatory
> variables.
>
> I want to estimate a fixed effect model with fixed effects for: firms and
time.
>
> My dependent variable is a binary one (0/1).
>
> How should I proceed in stata?
>
> I added (T-1) time dummies and estimate the model with clogit including firm
as
> the "group" variable. Therefore, what I did is the following:
>
> clogit y1 x1-xk i.t, group(firm)
>
> Does anyone know if it is correct to include time dummies when using clogit?
>
> How do I check for a firm specific effect?
>
> How do I assess the goodness of fit of the model?
>
>
> Many thanks
>
> Grazia
>
*
* 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/
