bug ? wilcox() (PR#243)

About this list Date view Thread view Subject view Author view Other groups

Subject: bug ? wilcox() (PR#243)
ATurpin@discoveriesinsight.org
Date: Sat 07 Aug 1999 - 10:09:31 EST


Message-Id: <199908070009.CAA02388@pubhealth.ku.dk>

> Hi,
>
> I am getting a strange result from the wilcox function in the code below.
> This code has worked for many other wilcox's, but not this particular one.
> The p-value printed should be much higher (consistent with the p-value
> for the t-test).
>
> I am using R Version 0.64.2 under Debian Linux.
>
> I hope it is not me being stupid...
>
> Regards,
>
> Andrew Turpin
> aturpin@discoveriesinsight.org
>
> --------------------------------------------------------------------------
> ----
>
> library(ctest)
> x <- vector(length=50)
> y <- vector(length=50)
> x <- c(0.624200,0.252100
> ,0.404500
> ,0.465000
> ,0.033500
> ,0.565500
> ,0.316000
> ,0.469800
> ,0.590200
> ,0.420200
> ,0.609800
> ,0.577800
> ,0.854200
> ,0.284800
> ,0.243600
> ,0.596100
> ,0.213200
> ,0.369800
> ,0.281400
> ,0.754500
> ,0.135200
> ,0.041900
> ,0.640900
> ,0.547000
> ,0.214300
> ,0.212000
> ,0.325200
> ,0.400700
> ,0.115400
> ,0.248800
> ,0.211500
> ,0.540200
> ,0.282100
> ,0.131700
> ,0.757500
> ,0.136500
> ,0.391500
> ,0.293500
> ,0.275400
> ,0.048700
> ,0.308400
> ,0.140900
> ,0.624100
> ,0.042600
> ,0.158100
> ,0.631300
> ,0.279200
> ,0.623100
> ,0.042600
> ,0.477200
> )
> y <- c(0.606500,0.252100
> ,0.404500
> ,0.465000
> ,0.033500
> ,0.565500
> ,0.316000
> ,0.469800
> ,0.591700
> ,0.420200
> ,0.608700
> ,0.577800
> ,0.854200
> ,0.284800
> ,0.244700
> ,0.596100
> ,0.213200
> ,0.369800
> ,0.281400
> ,0.754500
> ,0.135200
> ,0.041900
> ,0.640900
> ,0.683500
> ,0.214300
> ,0.192100
> ,0.304800
> ,0.400700
> ,0.115400
> ,0.248800
> ,0.211500
> ,0.540200
> ,0.282100
> ,0.131700
> ,0.757500
> ,0.137200
> ,0.391500
> ,0.293500
> ,0.275400
> ,0.048700
> ,0.308400
> ,0.140900
> ,0.624100
> ,0.042600
> ,0.158100
> ,0.631300
> ,0.282900
> ,0.620100
> ,0.030800
> ,0.471700
> )
> #
> # some functions for formatting output
> #
> f <- function (x) { formatC(x,digits=4,width=10,format="f") }
> mydiff<-function (x,y) {
> formatC((y-x)/y*100,digits=2,width=10,format="f")}
> line <- function (d,x,y) { cat(d,f(x),f(y),mydiff(x,y),"\n")}
>
> #
> # print basic stats
> #
> line("n ",length(x),length(y))
> line("Mean ",mean(x),mean(y))
> line("Median ",median(x),median(y))
> line("Var ",var(x),var(y))
>
> cat("\n")
>
> #
> # Perform basic tests
> #
> w <- wilcox.test(x,y,paired=TRUE)
> t <- t.test(x,y,paired=TRUE)
>
> cat("t-test p =", f(t$p.value),
> ifelse(t$p.value<0.05,"(Significant)","(Not significant)"), "\n")
> cat("Wilcoxon p =",
> f(w$p.value),ifelse(w$p.value<0.05,"(Significant)","(Not significant)"),
> "\n")
>

-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-.-
r-devel mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-devel-request@stat.math.ethz.ch
_._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._._


About this list Date view Thread view Subject view Author view Other groups

This archive was generated by hypermail 2b25 : Tue 04 Jan 2000 - 14:16:07 EST