Statalist


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

st: RE: t-test comparing the means of two samples in imputed datasets


From   "David Radwin" <dradwin@mprinc.com>
To   <statalist@hsphsun2.harvard.edu>
Subject   st: RE: t-test comparing the means of two samples in imputed datasets
Date   Wed, 4 Nov 2009 13:13:37 -0800 (PST)

Clara,

This is only a partial answer because I am not an expert on how MI works
in Stata, but the equivalent to -ttest var, by(dummy)- without MI would be
-regress var dummy- (your standard bivariate OLS regression). The
coefficient on dummy is the mean difference and the t-ratio and two-tailed
p-value are identical to those produced by -ttest-.

Perhaps another Statalister can confirm or deny that this equivalence
holds in the MI world as well.

David

--
David Radwin  // dradwin@mprinc.com
Research Associate
MPR Associates, Inc.
(510) 849-4942
http://mprinc.com


-----Original Message-----
From: owner-statalist@hsphsun2.harvard.edu
[mailto:owner-statalist@hsphsun2.harvard.edu] On Behalf Of Clara Barata
Sent: Wednesday, November 04, 2009 10:42 AM
To: statalist@hsphsun2.harvard.edu
Subject: st: t-test comparing the means of two samples in imputed datasets

Hey!
Any idea on how to apply a ttest to compare means in datasets imputed
with MI (stata 11)?
What would be the equivalent to: "ttest var  , by (dummy)" in the MI
world?

I searched through "mi estimate" and "mi test", but couldn't find
anything that matched.
Thanks, clara

*
*   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 1996–2014 StataCorp LP   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index