Stata The Stata listserver
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

Re: st: Negative Binomial Fixed Effects estimator


From   Paul Millar <paul.millar@shaw.ca>
To   statalist@hsphsun2.harvard.edu
Subject   Re: st: Negative Binomial Fixed Effects estimator
Date   Wed, 02 Mar 2005 22:49:38 -0700

Hello Anirban,

You can think of the "fixed effects" cross-sectional time series models like pooled (or "stacked") cross-sections, with a dummy variable for each unit of analysis. This makes them quite different than the more traditional kind of "fixed effect" model, which is comparing treatment effects.

Your model probably has a different age at each time point for each person. That is why it varies, and why it has an effect.

- Paul Millar
Sociology
University of Calgary


At 01:32 PM 02/03/2005, you wrote:

Hi,

I am running Stata 8 SE.

Does anyone know why xtnbreg, fe generates estimates for covariates that do not vary within group. For example, my outcome variable is y, which is a count variable and I have two period of data for every caseid.

. xtnbreg y age time, fe nolog i(caseid)
note: you are responsible for interpretation of non-count dep. variable
note: 6096 groups (12192 obs) dropped due to all zero outcomes

Conditional FE negative binomial regression Number of obs = 992
Group variable (i): caseid Number of groups = 496

Obs per group: min = 2
avg = 2.0
max = 2

Wald chi2(2) = 188.18
Log likelihood = -402.55541 Prob > chi2 = 0.0000

------------------------------------------------------------------------------
y | Coef. Std. Err. z P>|z| [95% Conf. Interval]
-------------+----------------------------------------------------------------
age | .0135565 .0216634 0.63 0.531 -.028903 .0560161
time | -1.505405 .1098068 -13.71 0.000 -1.720623 -1.290188
_cons | -2.399964 .5905975 -4.06 0.000 -3.557513 -1.242414
------------------------------------------------------------------------------

. sort caseid time

. count if caseid ==caseid[_n-1] & age !=age[_n-1]
0

Covariate age does not vary by period within caseid. What is the interpretation of the coefficient on age? Shouldn't age drop out in a fixed effects model?



Thanks,

Anirban


_______________________________________
Anirban Basu Ph.D.
Section of General Internal Medicine
Department of Medicine
University of Chicago
5841 S. Maryland Ave, MC-2007
Chicago IL 60637

Tel: +1 773 834 1796
Fax: +1 773 834 2238


NOTICE OF CONFIDENTIALITY - This material is intended for the use of the individual or entity to which it is addressed, and may contain information that is privileged, confidential and exempt from disclosure under applicable laws. If the reader of this material is not the intended recipient, you are hereby notified that any dissemination, distribution or copying of this communication is strictly prohibited. Please notify the sender of the error and destroy the Email you received.
*
*   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