Statalist The Stata Listserver

[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: RE: RE: GMM + dummy variable

From   "Austin Nichols" <>
Subject   Re: st: RE: RE: GMM + dummy variable
Date   Wed, 8 Nov 2006 11:39:07 -0500

You may want year dummies, too, and for that I recommend using -tab-
with a -gen- option:

ssc install ivreg2, replace
ssc install xtivreg2, replace
ssc install xtabond2, replace
use, clear
tsset id year
tab ye, gen(dy)
drop dy1
g highorlow=(ind==5 | ind==7) if !mi(ind)
xtivreg2 ys k n dy* (=indout) if hi==0, fe cl(id) gmm small
xtivreg2 ys k n dy* (=indout) if hi==1, fe cl(id) gmm small

Also read the help files for -xtivreg2- and -xtabond2- and the
references therein for descriptions of various good panel methods.
(The example above is not a good method, just illustrative of syntax.)

On 11/8/06, Nick Cox <> wrote:
Samuel already has his dummy, so I don't think that
is the issue. I don't think that mentioning GMM
ties this problem down precisely, any more than
saying that one is using maximum likelihood.


Maarten Buis

> I don't know about the gmm, but for the dummies have a look
> at -help xi-.

samuel allen

>  I need to compute two different regressions.
> This is my variables list:   company, year (from
> 1980 to 1990), id (1, 2, 3...), var1, var2... ,
> highorlow (it is an already created column of 1 and
> 0, a number for each firm).
> Could somebody write me command-examples using this
> dummy?
*   For searches and help try:

© Copyright 1996–2015 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index