<>
" e(v) won't but e(V) will, note the capital letter V."
It is the same as in the -return list- after -summarize-: You have to capitalize -r(Var)- to get it right (been bitten a thousand times by it...)
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected] [mailto:[email protected]] Im Auftrag von Maarten buis
Gesendet: Dienstag, 7. Juli 2009 22:48
An: [email protected]
Betreff: Re: st: Could Stata give the variance of the parameter estimates?
--- Crystal wrote:
> -matrix list e(v)- doesn't work
e(v) won't but e(V) will, note the capital letter V.
> I am going to use these var and covar to
> calculate the standard error for other factor (let's say SE(Mij)),
> according to a given fomular. But how could I tell if the Mij is
> significant at the 5% level, according to the calculated SE(Mij)?
The standard error is nothing other than the standard deviation
of the sampling distribution, so the answer depends on what you
know about the sampling distribution. A common answer is that
you think that the sampling distribution is Gaussian (normal),
in that case see: Buis, Maarten L. (2007) Stata tip 54: Where did
my p-values go? The Stata Journal, 7(4):584--586.
http://home.fsw.vu.nl/m.buis/wp/pvalue.html . But a Gaussian
sampling distribution is often not the appropriate sampling
distribution (think for example of chi square, F, and t test).
-- Maarten
-----------------------------------------
Maarten L. Buis
Institut fuer Soziologie
Universitaet Tuebingen
Wilhelmstrasse 36
72074 Tuebingen
Germany
http://home.fsw.vu.nl/m.buis/
-----------------------------------------
*
* 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/
