Statalist The Stata Listserver


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

st: Re: xtabond


From   "Rodrigo A. Alfaro" <ralfaro76@hotmail.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: Re: xtabond
Date   Sat, 19 Aug 2006 23:51:47 -0400

Bartlomiej,

You want to use Arellano-Bond procedure because in theory
you have a model like this (basic one)

y(i,t) = b*y(i,t-1) + a(i) +e(i,t)

a(i) is an unobservable cross-sectional unit effect. OK?

So far, I don't think there will be a serious problem to estimate
your dynamic binary model with -probit- or -logit- procedures. 
But becomes more complicated if you have a(i) in your model.

The solutions that I know are: -xtlogit- (fixed or random effects),
-xtprobit- (random effect) or -logit/probit- with dummies. 

Rodrigo.



----- Original Message ----- 
From: "Bartlomiej Wojcik" <bw202961@students.mimuw.edu.pl>
To: <statalist@hsphsun2.harvard.edu>
Sent: Saturday, August 19, 2006 7:14 PM
Subject: st: xtabond


dear stata users

my question is as follows. is xtabond procedure adequate for binary 
response variable? if not, then how can I model dynamic model with 
discrete dependent variable?

kind regards
b.wojcik
*
*   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/



© Copyright 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index