You cannot select more than 25 topics
Topics must start with a letter or number, can include dashes ('-') and can be up to 35 characters long.
40 lines
1.2 KiB
Plaintext
40 lines
1.2 KiB
Plaintext
9 months ago
|
{smcl}
|
||
|
|
{* 31jan2002}{...}
|
||
|
|
{hline}
|
||
|
|
help for {hi:xsampsi}
|
||
|
|
{hline}
|
||
|
|
|
||
|
|
{title:Sample size for AB/BA cross-over trials with continuous measures}
|
||
|
|
|
||
|
|
This program will compute the required sample size for a simple crossover
|
||
|
|
trial, that is AB/BA, or two-period two-treatment cross-over. It will
|
||
|
|
estimate the power for a given sample size, too.
|
||
|
|
|
||
|
|
You must specify:
|
||
|
|
alpha required significance level
|
||
|
|
beta 1-power
|
||
|
|
delta the difference between treatment arms
|
||
|
|
stddev the standard deviation of this difference
|
||
|
|
|
||
|
|
You can specify:
|
||
|
|
n different sample sizes, for which power will be computed
|
||
|
|
|
||
|
|
Requirements
|
||
|
|
You must install the package nct ({help nct}) for this program to work.
|
||
|
|
|
||
|
|
Comments
|
||
|
|
This program essentially duplicates step by step the method that is
|
||
|
|
given in Stephen Senn, "Cross-over trials in clinical research", 1993,
|
||
|
|
on page 217. It is almost the same as a one-sample t-test for the
|
||
|
|
difference between treatment, except that this method uses the non-central
|
||
|
|
t-distribution, which gives you more accurate answers.
|
||
|
|
|
||
|
|
The variables are assumed to be normally distributed.
|
||
|
|
|
||
|
|
|
||
|
|
See also:
|
||
|
|
{help sampsi}, {help studysi}
|
||
|
|
|
||
|
|
Written by Jan Brogger (jan.brogger@med.uib.no)
|
||
|
|
|