Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
st: AW: AW: One-sample ttests with estpost
From
"Martin Weiss" <[email protected]>
To
<[email protected]>
Subject
st: AW: AW: One-sample ttests with estpost
Date
Thu, 10 Jun 2010 16:02:56 +0200
<>
As in:
*************
sysuse auto, clear
keep price weight length trunk
eststo clear
qui foreach var of varlist _all{
//random samples via -if-
eststo, tit("`var'"): reg `var' if runiform()<.7
su `var',d
estadd local sd=round(r(sd),2)
}
esttab *, p /*
*/ sca("sd Standard Deviation") /*
*/ obslast sfmt(%9.0g)
*************
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected] [mailto:[email protected]] Im Auftrag von Martin Weiss
Gesendet: Donnerstag, 10. Juni 2010 15:06
An: [email protected]
Betreff: st: AW: One-sample ttests with estpost
<>
You may want to exploit the equivalence btw -ttest- and -reg-, though:
*************
sysuse auto, clear
ttest mpg==0
reg mpg
*************
HTH
Martin
-----Ursprüngliche Nachricht-----
Von: [email protected] [mailto:[email protected]] Im Auftrag von Florian Szuecs
Gesendet: Donnerstag, 10. Juni 2010 11:47
An: [email protected]
Betreff: st: One-sample ttests with estpost
Dear Statalisters,
I'm trying to automatically generate a large table containing means, s.d.,
observations and one-sample ttests (of the form ttest x=0) for multiple
variables across different subsamples. My approach was to use estpost
ttest for each subsample and then join the results using esttab. This
would allow me to display the ttest-results as significance stars next to
the means.
Estpost, however, only supports two-mean ttests and simply setting all obs
out of the current subsample to zero for the ttest meddles with the
degrees of freedom. In a similar matter, statalisters suggested using
either a matrix or the parmest command to obtain the desired results, but
I failed to adapt this to my problem.
Any ideas on how to get estpost to do what I want it to do?
Thanks in advance,
Florian
*
* 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/
*
* 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/