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: Hausman-Taylor AR(1) estimator


From   "Schaffer, Mark E" <M.E.Schaffer@hw.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: Hausman-Taylor AR(1) estimator
Date   Thu, 6 Sep 2012 20:45:27 +0100

Koray,

You'll need to show us the output with the error message about the lack
of full rank etc. before we can offer advice about what it means and how
to deal with it.

You can use xtoverid to get cluster-robust SEs; this is a way of
addressing the within-panel autocorrelation.  This post briefly
describes how:

http://www.stata.com/statalist/archive/2011-03/msg00414.html

--Mark

> -----Original Message-----
> From: owner-statalist@hsphsun2.harvard.edu [mailto:owner-
> statalist@hsphsun2.harvard.edu] On Behalf Of KORAY ERCIHAN
> Sent: 06 September 2012 17:30
> To: statalist@hsphsun2.harvard.edu
> Subject: st: Hausman-Taylor AR(1) estimator
> 
> Dear Statalist,
> 
> The issues related to Hausman and Taylor estimator have been indicated
in
> Statalist but I couldn't find a solution for my problem.
> I am using panel data including 168 bilateral trade relations under
1993-2007
> years. I want to estimate Hausman-Taylor estimator (HT). Since I find
> autocorrelation in my data, I want to estimate HT AR(1) too.
> 
> I estimated two HT models for my variables. One of them (HT1) is
having 3
> variables in endog part and the other (HT2) has 4 variables in endog
part. I
> have 7 explanatory variables at hand and I'm using time dummies as
well. The
> overidentification test results; 0.11 with p-value: 0.9903 and 0.12
with p-
> value: 0.9983 for HT1 and HT2 respectively. However, I obtained these
test
> results after commanding "xtoverid, noi" but when I use bootstrap
option for
> xthtaylor, it gives an error about the lack of full rank for
covariance matrix
> then I can't get the overidentification test result.
> 
> Do you think I can rely on my overidentification results since their
p-values
> are very high? and how can I solve the autocorrelation problem?
> 
> I will very grateful if you share your knowledge.
> 
> Thanks
> 
> Koray
> 
> *
> *   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/


-- 
Heriot-Watt University is the Sunday Times
Scottish University of the Year 2011-2012

We invite research leaders and ambitious early career researchers to 
join us in leading and driving research in key inter-disciplinary themes. 
Please see www.hw.ac.uk/researchleaders for further information and how
to apply.

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