Statalist


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: AW: AW: ereturn list with xtivreg2


From   "Martin Weiss" <[email protected]>
To   <[email protected]>
Subject   st: AW: AW: ereturn list with xtivreg2
Date   Tue, 10 Mar 2009 16:28:26 +0100

<> 

The p-values seem to be calculated just the way you would expect them to be
calculated, i.e. in M. Buis'
http://www.stata-journal.com/article.html?article=st0137


*************
use http://fmwww.bc.edu/ec-p/data/macro/abdata.dta, clear
qui tsset id year
xtivreg2 ys k (n=l2.n l3.n), fe small
mata
mata clear
b=st_matrix("e(b)")'
se=diagonal(st_matrix("e(V)"):^0.5)
df = st_numscalar("e(df_r)")
t = b :/ se
t
p=2*ttail(df, abs(t))
p
end
*************



HTH
Martin


-----Ursprüngliche Nachricht-----
Von: [email protected]
[mailto:[email protected]] Im Auftrag von Davide Cantoni
Gesendet: Dienstag, 10. März 2009 15:44
An: [email protected]
Betreff: st: AW: ereturn list with xtivreg2

Martin, thanks for your reply.

Of course, I should have been more precise. I just mean the p-values
for the coefficient estimates. (Fifth column from the left in the main
output table, if I have to be even more precise)

Well, the issue is precisely that there is *no* "e(df_r)" in the
results saved in e() by -xtivreg2-.... So I would love to know what
all these results in e() mean.

Best, Davide



> <>
>
> What does "the p-value" mean? I see a couple of them in standard
-xtivreg2-
> output, and the returned value " e(df_r)" certainly fits your description.
> You have to clarify your question, I am afraid...
>
>
>
> HTH
> Martin
>
> -----Ursprüngliche Nachricht-----
> Von: [email protected]
> [mailto:[email protected]] Im Auftrag von Davide
Cantoni
> Gesendet: Dienstag, 10. März 2009 04:36
> An: [email protected]
> Betreff: st: ereturn list with xtivreg2
>
> Dear Statalisters,
>
> I have a very simple question: is there a way to find an explanation
> of what the various results saved in e() by -xtivreg2- actually
> correspond to? The official help file does not describe those, and
> while some are pretty standard, like e(r2_a), others seem to be
> specific to this command and their abbreviations are not immediately
> intellegible. (Specifically, I am searching for the residual degrees
> of freedom that are used when computing the p-value).
>
> Thanks,
>
> Davide

*
*   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index