From: Grathwohl, Dominik, LAUSANNE, NRC-BAS <dominik.grathwohl_at_rdls.nestle.com>

Date: Tue 25 Jul 2006 - 20:48:11 EST

sv4 <- summary(v4)

language R

R-help@stat.math.ethz.ch mailing list

https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Tue Jul 25 21:03:25 2006

Date: Tue 25 Jul 2006 - 20:48:11 EST

# 1. Method: Adjustment with library(multcomp)

library(nlme)

library(multcomp)

# the correct degrees of freedom covm=cov4, cmatrix=cm4, conf.level=0.95)

sv4 <- summary(v4)

*# 2. Method: I found in Handbook of Statistics Vol 13, p.616,
*

# same can be found in http://home.clara.net/sisa/bonhlp.htm

# Bonferroni on correlated outcomes:

raw.p <- sv4$p.value.raw

co4 <- cor(df$y[df$v==0],df$y[df$v==1]) rho <- mean(c(1,co4,co4,1)) pai <- 1-(1-raw.p)^2^(1-rho)

# The results of two methods are presented in the following lines:

out <- cbind(raw.p, sv4$p.value.bon, sv4$p.value.adj, pai)
colnames(out) <- c("raw.p", "bon.p", "multcomp.p", "bon.cor.p")
out

As you can see there are quite big differences between the two ways adjusting for multiple tests on repeated measurements. I guess that the multcomp library is not appropriate for this kind of hypotheses. However I could not find an explanation in the help files. May be one of the experts can point me in the right direction?

Kind regards,

Dominik

platform i386-pc-mingw32

arch i386 os mingw32 system i386, mingw32 status major 2 minor 2.1 year 2005 month 12 day 20 svn rev 36812

language R

[[alternative HTML version deleted]]

R-help@stat.math.ethz.ch mailing list

https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/posting-guide.html and provide commented, minimal, self-contained, reproducible code. Received on Tue Jul 25 21:03:25 2006

Archive maintained by Robert King, hosted by
the discipline of
statistics at the
University of Newcastle,
Australia.

Archive generated by hypermail 2.1.8, at Thu 03 Aug 2006 - 08:16:52 EST.

*
Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help.
Please read the posting
guide before posting to the list.
*