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

st: RE: ttest-thanks


From   "Vidya Mahambare" <MahambareV@Cardiff.ac.uk>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: ttest-thanks
Date   Thu, 04 Dec 2003 13:53:54 +0000

Yes, agreed. Many thanks for the reference and the comments.

Vidya


>>> n.j.cox@durham.ac.uk 12/04/03 01:14pm >>>
You indicated later that you had solved 
the question of how to do this in Stata. 
Irrespective of that, the t test is based
on the assumption that observations 
are independent, which in most panel 
situations seems unlikely; otherwise,
your P-values are seriously inaccurate. 
That doesn't make analysis impossible, but it does mean 
that the autocorrelation is needed to 
interpret the t test properly. 

Among many other sources, Rupert G. 
Miller's "Beyond ANOVA" is good on 
this kind of problem. 

Nick 
n.j.cox@durham.ac.uk 

Vidya Mahambare

> I am having some problem using ttest. I have a panel data - 
> 10 time periods,
> 2000+ firms from 13 sectors and 3 ownership categories.
> I am interested in testing whether the means of a variable 
> (efficiency
> indicator) are different for the ownership pairs during a 
> particular time period
> for each sector.
> Eg; time = 1 to 4, ttest efficiency if owner =1 and owner 
> =3 for sector 1, then
> sector 2 and so on
> Time = 5 to 8, ttest efficiency if owner =1 and owner =3 
> for sector 1, then
> sector 2 and so on
> 
> Should I temporality drop observations in all sectors 
> except one and then do
> ttest? I have tried several different ways, but without success. 
> 

*
*   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/
*
*   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