Raw Message

Message-ID: <20111107115918.1092.F023FAF3@mpi-bpc.mpg.de>
Date: 2011-11-07T10:59:19Z
From: Holger Taschenberger
Subject: one sample Wilcoxon test using 'coin'

Hi,

        I'm trying to use the package 'coin' to run a one sample Wilcoxon
test equivalent to this:

x1<-c(1,3.5,2.1,4,1.5,5)
wilcox.test(x1, mu=2, exact=TRUE)

I assume that I can do this like so:

x2<-rep(2,length(x1))
wilcoxsign_test(x1 ~ x2,distribution = exact())

But I'm not sure if this is really the correct way.
Can someone please advise?

(BTW: The reason to use 'coin' is it's ability to compute exact p-values
even in the presence of ties in the ranks.)

Thanks a lot,
        Holger