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: meta-analysis of proprotions with Freeman-Tukey arcsin transformation - how to transform back?
From
Michael Grosse <[email protected]>
To
[email protected]
Subject
st: meta-analysis of proprotions with Freeman-Tukey arcsin transformation - how to transform back?
Date
Tue, 18 Feb 2014 21:14:34 +0100
Dear all,
I need to perform a meta-analysis of proprotions - I´m using a
Freeman-Tukey arcsin transformation to stabilize variances.
I performed transformation as follows:
gen pTransformed = asin(sqrt(fraction/(all+1))) +
asin(sqrt((fraction+1)/(all+1)))
gen SEpTransformed = sqrt(1/(all+1))
metan pTransformed SEpTransformed, random
Meta-Analysis is now performed using the transformed values. However I
think it is not a good idea to report the results in transformed
units. Is there a possibility to generate results from the
back-transformed values? If yes, how?
Thanks,
Michael
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/faqs/resources/statalist-faq/
* http://www.ats.ucla.edu/stat/stata/