Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

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

st: Repost from a few months ago: Sample size estimation using GEEs

From   "Mark Marshall" <>
To   <>
Subject   st: Repost from a few months ago: Sample size estimation using GEEs
Date   Sun, 17 Apr 2011 07:47:59 +1200

I wish to determine the optimal sample size for a clinical trial of
intervention versus placebo where the outcomes are repeated measures. 

I can do this in Stata with a repeated measures ANCOVA using the sampsi
command. However, this looks a little odd in the grant application when I
then propose GEEs or GLMs in the statistical plan for data analysis.

There is a package in R and Splus called longpower which codes Liu and
Liang's formula for linear GEE models (page 941). I gather there is also a
package is SAS which codes Rochon's formula. There are probably many more
variations on these approaches published in different papers, but I gather
these are the most popular. 

Is there a user-written Stata ado file that codes any of these


Mark Marshall

Liu G, Liang KY. Sample size calculations for studies with correlated
observations. Biometrics. 1997;53:937-47.

Rochon, J. Application of GEE procedures for sample size calculations in
repeated measures experiments.  Statistics in Medicine. 1998;17, 1643-1658.

*   For searches and help try:

© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index