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: RE: hausman test results


From   "Justina Fischer" <JAVFischer@gmx.de>
To   statalist@hsphsun2.harvard.edu
Subject   Re: RE: st: RE: hausman test results
Date   Sat, 12 Feb 2011 11:08:58 +0100

well,, read what stata says: you are violating one of the assumptions of the Hausman test so that the calculated value has no meaning. 

without providing the code in the do-file it is difficult to see where this might come from. 

Justina
-------- Original-Nachricht --------
> Datum: Sat, 12 Feb 2011 10:54:02 +0100
> Von: Jan Lid <hhb010@hotmail.com>
> An: statalist@hsphsun2.harvard.edu
> Betreff: RE: st: RE: hausman test results

> you were right. I did do something wrong as my results are different this
> time. But i still don't know what this mean. Is a random effect model valid
> or nor valid? Says h0 efficient? So random effect valid?
> 
> Help very much appreciated
> 
> ---- Coefficients ----
> |           (b)         (B)        (b-B)     sqrt(diag(V_b-V_B))
> |             re         fe       Difference      S.E.
> -------------+----------------------------------------------------------------
> is |    .0027526      -.0002958    .0030484    .
> e_n |   -.0676737    -.0452386   -.0224351     .
> lnbtm_n | -.0189196    -.0295569 .0106373     .
> lnmve |  .0003718    -.0020215   .0023934     .
> nege |   .0230284    .0271639    -.0041354     .
> roe |   -.1102508    -.1590028   .048752     .
> posacc | .0300681    .0251275   .0049406     .
> negacc | .0330714    .0297334   .003338      .
> assg |   -.0025524   -.003037   .0004845     .
> nod |   -.0077827   -.0047668   -.0030159    .
> dtb |   .0600797     .0627425   -.0026628    .
> d1997 | -.0086867   -.0100742   .0013876   .0011561
> d1998 | -.0140629   -.0170658   .0030029    .0000744
> d1999 | -.0008133   -.0049143   .004101      .
> d2000 | -.0077408   -.0097045    .0019636     .
> d2001 | -.0234343   -.0232871   -.0001472     .
> d2002 | -.0116084   -.0147064   .0030979      .
> d2003 | .0045587   .0039078      .000651      .
> d2004 | .002967    .0033895     -.0004225     .
> d2005 | .0034413   .0058742     -.0024328     .
> d2006 | .0036623   .0052834     -.0016211     .
> d2007 | .0006761   .0029451     -.002269      .
> d2008 | -.0116638   -.0092098   -.002454      .
> d2009 | -.0007347   .0024991    -.0032338     .
> ------------------------------------------------------------------------------
> b = consistent under Ho and Ha; obtained from xtreg
> B = inconsistent under Ha, efficient under Ho; obtained from xtreg
> Test: Ho: difference in coefficients not systematic
> chi2(24) = (b-B)'[(V_b-V_B)^(-1)](b-B)
> = -91.23 chi2<0 ==> model fitted on these
> data fails to meet the asymptotic
> assumptions of the Hausman test;
> see suest for a generalized test
> 
> 
> 
> 
> > Date: Fri, 11 Feb 2011 21:31:27 +0100
> > From: JAVFischer@gmx.de
> > Subject: Re: st: RE: hausman test results
> > To: statalist@hsphsun2.harvard.edu
> > 
> > Hi Lid,
> > 
> > Eric is right. Your coefficients for the FE and RE model are identical.
> This is extremely unlikely to be the case ...
> > 
> > Justina
> > -------- Original-Nachricht --------
> > > Datum: Fri, 11 Feb 2011 19:19:33 +0100
> > > Von: DE SOUZA Eric <eric.de_souza@coleurope.eu>
> > > An: "statalist@hsphsun2.harvard.edu" <statalist@hsphsun2.harvard.edu>
> > > Betreff: st: RE: hausman test results
> > 
> > > The results indicate that you did something wrong. The kind of results
> > > you get is what would happen if did the following:
> > > webuse grunfeld
> > > xtreg invest mvalue kstock, fe
> > > estimates store fef
> > > xtreg invest mvalue kstock, fe
> > > estimates store ref
> > > hausman fef ref
> > > 
> > > Did you get the following message before your test output?
> > > 
> > > Note: the rank of the differenced variance matrix (0) does not equal
> the
> > > number of coefficients being tested (2); be
> > > sure this is what you expect, or there may be problems computing
> > > the test. Examine the output of your
> > > estimators for anything unexpected and possibly consider scaling
> > > your variables so that the coefficients are on
> > > a similar scale.
> > > 
> > > 
> > > Eric de Souza
> > > College of Europe
> > > Brugge (Bruges), Belgium
> > > http://www.coleurope.eu
> > > 
> > > -----Original Message-----
> > > From: owner-statalist@hsphsun2.harvard.edu
> > > [mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Jan Lid
> > > Sent: 11 February 2011 18:32
> > > To: statalist@hsphsun2.harvard.edu
> > > Subject: st: hausman test results
> > > 
> > > 
> > > 
> > > 
> > > 
> > > 
> > > Hi,
> > > 
> > > I have a prroblem interpreting the hausman test. My difference is zero
> > > everywhere which valids my use of random effect model. What i do not
> > > understand however is that my hausman value result is zero. with a
> probability of
> > > zero. This would suggest systematic difference but the table below
> tells me
> > > otherwise. How do i interpret this? Thx for the help
> > > 
> > > Jan Lid
> > > 
> > > 
> > > ---- Coefficients ----
> > > | (b) (B) (b-B) sqrt(diag(V_b-V_B))
> > > | fixed random Difference S.E.
> > >
> -------------+----------------------------------------------------------
> > > -------------+------
> > > e_n | -.0992885 -.0992885 0 0
> > > lnbtm_n | -.0044924 -.0044924 0 0
> > > lnmve | .0237916 .0237916 0 0
> > > nege | .0231327 .0231327 0 0
> > > roe | .0245178 .0245178 0 0
> > > acc | .0664807 .0664807 0 0
> > > | .0450786 .0450786 0 0
> > > assg | .195635 .195635 0 0
> > > nod | .1440675 .1440675 0 0
> > > dtb | .2710841 .2710841 0 0
> > > d1997 | -.0355099 -.0355099 0 0
> > > d1998 | -.0145913 -.0145913 0 0
> > > d1999 | .0119304 .0119304 0 0
> > > d2000 | -.0550707 -.0550707 0 0
> > > d2001 | -.1669835 -.1669835 0 0
> > > d2002 | .0074249 .0074249 0 0
> > > d2003 | .1367043 .1367043 0 0
> > > d2004 | .1281495 .1281495 0 0
> > > d2005 | .0103888 .0103888 0 0
> > > d2006 | -.0578083 -.0578083 0 0
> > > d2007 | -.020043 -.020043 0 0
> > > d2008 | -.0785053 -.0785053 0 0
> > > d2009 | -.0668654 -.0668654 0 0
> > >
> ------------------------------------------------------------------------------
> > > b = consdstent under Ho and Ha; obtained from xtreg B = inconsdstent
> under
> > > Ha, efficient under Ho; obtained from regress
> > > 
> > > Test: Ho: difference in coefficients not systematic
> > > 
> > > chi2(0) = (b-B)'[(V_b-V_B)^(-1)](b-B)
> > > = 0.00
> > > Prob>chi2 = .
> > > (V_b-V_B ds not positive definite) 
> > > *
> > > * 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/
> > > 
> > > *
> > > * 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/
> > 
> > -- 
> > Justina AV Fischer, PhD
> > Senior Researcher
> > Faculty of Economics
> > University of Mannheim
> > 
> > homepage: http://www.justinaavfischer.de/
> > e-mail: javfischer@gmx.de
> > papers: http://ideas.repec.org/e/pfi55.html
> > 
> > 
> > *
> > * 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/ 		 	   		  
> *
> *   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/

-- 
Justina AV Fischer, PhD
Senior Researcher
Faculty of Economics
University of Mannheim

homepage: http://www.justinaavfischer.de/
e-mail: javfischer@gmx.de
papers: http://ideas.repec.org/e/pfi55.html


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