Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re:Re: st: Hausman test after xtnbreg and alternatives?


From   Christopher Baum <[email protected]>
To   "[email protected]" <[email protected]>
Subject   Re:Re: st: Hausman test after xtnbreg and alternatives?
Date   Tue, 12 Feb 2013 14:05:35 +0000

<>

NO!!! As I explained in a recent post, you cannot arbitrarily choose the order of arguments to -hausman-:

http://hsphsun3.harvard.edu/cgi-bin/lwgate/STATALIST/archives/statalist.1302/date/article-62.html

David is right in saying if you do, the negative sign will disappear, but then you are misusing the command, and its result is meaningless garbage.

Note that in xtnbreg, the fixed vs random effect does not enter the 'xb' part of the model, as it does in linear
regression, but refers to the overdispersion parameter varying across panel units. Whether there is any way of testing the additional
overidentifying restrictions a la -xtoverid- is not clear to me.

On Feb 12, 2013, at 2:33 AM, David wrote:

> Date: Mon, 11 Feb 2013 14:11:36 -0500
> From: David Greenberg <[email protected]>
> Subject: Re: st: Hausman test after xtnbreg and alternatives?
> 
> Reverse the order in which the two models are introduced, and the
> negative sign will disappear. David Greenberg, Sociology Department,
> New York University
> 
> On Mon, Feb 11, 2013 at 1:23 PM, Heewon Chae <[email protected]> wrote:
>> Hi All,
>> 
>> I am using xtnbreg and I want to decide whether to choose fixed or
>> random effects.
>> 
>> When I run
>> xtnbreg .... , fe
>> estimates store fixed
>> xtnbreg .... , re
>> estimates store random
>> hausman fixed random
>> 
>> I get this result:
>> 
>> Test:  Ho:  difference in coefficients not systematic
>> 
>>                 chi2(12) = (b-B)'[(V_b-V_B)^(-1)](b-B)
>>                          =   -68.68    chi2<0 ==> model fitted on these
>>                                        data fails to meet the asymptotic
>>                                        assumptions of the Hausman test;
>>                                        see suest for a generalized test
>> 
>> What other commands or approaches I can use to check which one to use?
>> Unfortunately, the alternative to hausman proposed in this list,
>> xtoverid, does not work for xtnbreg. I would be very grateful for any
>> comments or suggestions that helped me to decide whether fixed or
>> random effects are appropriate in my case.



Kit Baum   |   Boston College Economics & DIW Berlin   |   http://ideas.repec.org/e/pba1.html
                             An Introduction to Stata Programming  |   http://www.stata-press.com/books/isp.html
  An Introduction to Modern Econometrics Using Stata  |   http://www.stata-press.com/books/imeus.html
                                                                                                   | http://www.crup.com.cn/Item/111779.aspx	


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index