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

Re: Fwd: st: tsset error in panel estimation


From   Mark Schaffer <M.E.Schaffer@hw.ac.uk>
To   statalist@hsphsun2.harvard.edu, armen@uni-hohenheim.de
Subject   Re: Fwd: st: tsset error in panel estimation
Date   Sat, 26 Feb 2005 18:24:29 +0000 (GMT)

Arem,

I'll have a go at answering a couple of your questions...

Quoting armen@uni-hohenheim.de:

> Dear all,
> 
> I have a balanced panel data. I used xtgls with heteroscedasticity
> and psar1 
> options specifying first iis district and tis year. That worked
> perfectly.
> However, when I went estimating my model with Beck and Katz's
> panel-corrected 
> standard errors, i.e. applied xtpcse I got an error message saying I
> have to 
> tsset! The same occurs when I try to test for cross-district
> correlation and 
> for homoscadsticity using xttest2 and xttest3 respectively?!
> 
> What am I doing wrong? 

I don't think there is anything wrong at all.  -help xtpcse- indicates that
your data have to be -tsset-; -help xtgls- does not specify this
requirement.  All this means is that the code inside the former requires
-tsset- to work, and the code inside the latter doesn't.  If you just
-tsset- your data, everything should be fine.

> Using the opportunity may be a two more questions? 
> 
> I would like namely to calculate elasticities. Does that work for
> panel 
> (N=235, T=29) the usual way with mfx, eyex?
> 
> Finally, what I assume with my IVs that one or two are actually
> endogenous.
> I would have applied 3SLS in a "normal" (non-panel) setting to
> address 
> endogeniety. Is there any way to do "3SLS" for xt?
> I could not find
> anything 
> like that in the literature. May be that is too dumn what I think?
> Actually, intuitively I think I take would-be-3sls structure,
> estimate each of 
> them SEPARATELY with xt

You would do this using xtivreg, no?  And if you want to go on to estimate
the equations as a system, you could put the data into mean-deviation form
"by hand" (e.g., using Ben Jann's -center-), and then use -reg3-.  The SEs
would need a dof correction, but that's not hard to do.

Hope this helps.

--Mark

> (the first equation being the principal) and
> then put 
> the results next to each other and treat them as a "real" 3SLS.
> Figuring out 
> elasticities I make a distinction in so-called direct and indirect
> effects of 
> my RHS variables on the main DV?!!
> 
> Any hint will be much appreciated. 
> 
> Thanks in advance.
> 
> Armen
> Doctorate student
> University of Hohenheim
> Germany
> 
> *
> *   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/
> 
> ----- Ende der weitergeleiteten Nachricht -----
> 
> 
> *
> *   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/
> 



Prof. Mark Schaffer
Director, CERT
Department of Economics
School of Management & Languages
Heriot-Watt University, Edinburgh EH14 4AS
tel +44-131-451-3494 / fax +44-131-451-3294
email: m.e.schaffer@hw.ac.uk
web: http://www.sml.hw.ac.uk/ecomes

-------------------------------------------------------------------
DISCLAIMER

This message is subject to http://www.hw.ac.uk/disclaim.htm 

-------------------------------------------------------------------

*
*   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