[R] Jonckheere-Terpstra test

[RAVI VARADHAN]

Hi Marie,

Here is a function that I wrote based on the "JT.test " function in package "SAGx",  to account for missing values and to order based on the groups. 

JT.fun <- function(y, g) {
 nas <- is.na(y) | is.na(g)
 ord <- order(g[!nas])
 d <- y[!nas][ord]
 g <- g[!nas][ord]
 return (JT.test(data=t(as.matrix(d)),class=g) )
 }

Here is an example showing how to use it:

> library(SAGx)
>  y <- rnorm(30, mean=rep( c(0,1,2),  ea=10) )
>  grp <- rep(1:3, ea=10)
>  JT.fun( y , grp)

This will give you a two-sided p-value.  If your alternative is one-sided then you divide this p-value by 2.

Hope this helps,
Ravi.



----- Original Message -----
From: Marie Kraska <kraskmf at auburn.edu>
Date: Saturday, March 25, 2006 12:35 pm
Subject: Re: [R] Jonckheere-Terpstra test

> Hello, Can anyone tell me how to write the R script to get the P-value
> for the Jonckheere-Terpstra (JT) test? Following is the code I am 
> using.I cannot get the P-value to print out. Any help would be 
> appreciated.Thanks. Marie
> 
> 
> jt <- function(x, alpha=.05){
>        if(is.list(x) == 0) stop("Please enter a list of vectors.")
>        k <- length(x)
>        us <- matrix(0,k,k)
>        for(i in 1:(k-1)){
>                for(j in (i+1):k){
>                        us[i,j] <- wilcox.test(x[[j]],
> x[[i]])$statistic
>                }
>        }
>        sum(us)
> }
> 
> 
> con <- c(40,35,38,43,44,41)
> gpb <- c(38,40,47,44,40,42)
> gpc <- c(48,40,45,43,46,44)
> 
> jt(list(con,gpb,gpc))
> 
> ______________________________________________
> R-help at 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