Bookmark and Share

Notice: On March 31, it was announced that Statalist is moving from an email list to a forum. The old list will shut down on April 23, and its replacement, statalist.org is already up and running.


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

st: RE: generalized test alternative to Hausman


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: generalized test alternative to Hausman
Date   Tue, 6 Sep 2011 16:29:28 +0100

Pasquale,

A couple of thoughts...

1.  It looks like you're using the old syntax of the -hausman- command.  Is this because you have an old version of Stata?  FIY, when posting to Statalist, you should normally say which version of you are using (see the FAQ, http://www.stata.com/support/faqs/res/statalist.html#stata).

2.  The Hausman test stat can sometimes be <0 in finite samples because different estimates of the error variance are being used in V_b and V_B.  You can force the same variance to be used in both with either the sigmamore or sigmaless options (depending on the version of Stata you have, hence the importance of telling us - see #1).

HTH,
Mark

> -----Original Message-----
> From: Pasquale Pazienza [mailto:p.pazienza@unifg.it] 
> Sent: Monday, September 05, 2011 8:52 PM
> To: statalist@hsphsun2.harvard.edu
> Cc: Schaffer, Mark E
> Subject: generalized test alternative to Hausman
> 
> Dear Sir/Madam,
> 
> in dealing with my panel-data analysis I found a problem when 
> running the Hausman test 
> for the choice between RE and FE. The outcome is:
> 
> chi2(8) = (b-B)'[(V_b-V_B)^(-1)](b-B)= -25.37
> chi2<0 ==> model fitted on these data fails to meet the 
> asymptotic assumptions of the 
> Hausman test; see suest for a generalized test
> 
> In the various topics treated in STATALIST, I found a reply 
> dated 2 October 2003 to 
> a question posted by Joana Quina and representing a similar problem.
> 
> Following the suggested command and procedure (I report here 
> below) I am not able to 
> generate any useful results, since what I get is just the 
> indication of 
> <<invalid 'colnumb'>> when I run command no. 11 (here below), 
> which - of course - does 
> not allow me to run command 12. 
> 
> More specifically, according to the given hints I use the 
> following commands:
> 
> 1) local id my id
> 
> 2) local depvar Ln_CO2sct
> 
> 3) local varlist Ln_GDPsct_pw Ln_GDPsct_pw2 Ln_FLWsct_pGDP 
> Ln_SCTrel_2 Ln_GCF_pw 
> Ln_edu Ln_MKTopn_2 Ln_1_CRpr_GDP_1
> 
> 4) xtreg `depvar' `varlist', re
> 
> 5) hausman, save
> 
> 6) xtreg `depvar' `varlist', fe
> 
> 7) hausman, less
> 
> 8) tokenize `varlist'
> 
> 9) local i 1 while "``i''" != "" {qui by `id':  gen double 
> mean`i' = sum(``i'') / _n 
> qui by `id':  replace mean`i' = mean`i'[_n] qui by `id':  gen 
> double diff`i' = ``i'' - 
> mean`i' local newlist `newlist' mean`i' diff`i' local i = `i' + 1}
> 
> 10) xtreg `depvar' `newlist', re
> 
> 11) qui test mean1 = mean1 , notest /* clear test */ local i 
> 2 while "``i''" != "" {if 
> `b'[1, colnumb (`b', "mean`i'")] != 0 & /**/ `b'[1, colnumb 
> (`b', "diff`i'")] != 0 
> {qui test mean`i' = diff`i', accum notest} local i = `i' + 1}
> 
> 12) test
> 
> I would really appreciate to receive any help to solve this problem.
> Thanks in advance.
> Regards.
> 
> Pasquale
> 
> **************************************************************
> *************************
> WORK ADDRESSES:
> 
> (in Italy)                                (in UK)
> Università degli Studi di Foggia          Department of 
> Economics and Finance 
> Facoltà di Economia (DSEMS)               Durham University 
> Largo Papa Giovanni Paolo II n. 1         23/26 Old Elvet  
> 71100 Foggia (Italy)                      Durham DH1 3HY
> E-mail: <p.pazienza@unifg.it>             E-mail: 
> <pasquale.pazienza@durham.ac.uk>
> 
> 5 x 1000 - Campagna di Raccolta fondi 2011 a favore 
> dell'Universita' degli Studi di Foggia
> Sostieni la Ricerca Scientifica per il futuro della nostra 
> societa', dei nostri giovani, della scienza. Scegli di donare 
> il 5 x 1000 all'Universita' degli Studi di Foggia, nella tua 
> dichiarazione dei redditi firma e inserisci il codice fiscale 
> 94045260711
> 


-- 
Heriot-Watt University is a Scottish charity
registered under charity number SC000278.


*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   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   |   Site index