R help - Oct 2011 - one sample Wilcoxon test using 'coin'

Hi,

R allows me to run a one sample Wilcoxon test like this:
wilcox.test(c(1,3.5,2.1,4,1.5,5), mu=2, exact=TRUE)

The function 'wilcoxsign_test' from the package 'coin' should (I
suppose) be able to calculate exact p values even if there are ties in
the ranks. However, I couldn't find information on how to run a one
sample test using 'wilcoxsign_test' like in the example above.

Can anybody help?

        Thanks,
                Holger

The wilcoxon test in coin are meant only for "Testing the equality of the
distributions of a numeric response in *two or more* independent groups against
shift alternatives" whereas the wilcoxon test in the base pacakge
"Performs *one-* and two-sample Wilcoxon tests on vectors of data; the
latter is also known as ‘Mann-Whitney’ test."   So, wilcoxon_test and
wilcoxonsign_test in coin will not work for one-sample.  Why not just use
wilcox.test?
 
Mark J. Lamias


________________________________
From: Holger Taschenberger <Holger.Taschenberger@mpi-bpc.mpg.de>
To: r-help@r-project.org
Sent: Monday, October 31, 2011 11:12 AM
Subject: [R] one sample Wilcoxon test using 'coin'

Hi,

R allows me to run a one sample Wilcoxon test like this:
wilcox.test(c(1,3.5,2.1,4,1.5,5), mu=2, exact=TRUE)

The function 'wilcoxsign_test' from the package 'coin' should (I
suppose) be able to calculate exact p values even if there are ties in
the ranks. However, I couldn't find information on how to run a one
sample test using 'wilcoxsign_test' like in the example above.

Can anybody help?

        Thanks,
                Holger

______________________________________________
R-help@r-project.org 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.
	[[alternative HTML version deleted]]

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