This concerns the contributed package "concord". Sorry to bother
you with it, but my attempt to contact the author/maintainer
failed (see below). Perhaps you can forward it, or let me
know where to send it.
Regards, Rob Kushler
------------------------------------------------------
This is the Postfix program at host tak.itd.uts.edu.au.
I'm sorry to have to inform you that your message could not be
be delivered to one or more recipients. It's attached below.
For further assistance, please send mail to <postmaster>
If you do so, please include this problem report. You can
delete your own text from the attached returned message.
The Postfix program
<jim.lemon at uts.edu.au>: host 138.25.243.40[138.25.243.40] said: 550
5.1.6
recipient no longer on server: jim.lemon at uts.edu.au (in reply to RCPT TO
command)
Reporting-MTA: dns; tak.itd.uts.edu.au
X-Postfix-Queue-ID: 37A885D8027
X-Postfix-Sender: rfc822; kushler at oakland.edu
Arrival-Date: Fri, 28 Oct 2005 09:01:17 +1000 (EST)
Final-Recipient: rfc822; jim.lemon at uts.edu.au
Action: failed
Status: 5.0.0
Diagnostic-Code: X-Postfix; host 138.25.243.40[138.25.243.40] said: 550 5.1.6
recipient no longer on server: jim.lemon at uts.edu.au (in reply to RCPT TO
command)
Subject:
Page test in "concord" package
From:
Robert Kushler <kushler at oakland.edu>
Date:
Thu, 27 Oct 2005 19:01:01 -0400
To:
jim.lemon at uts.edu.au
I believe there is a bug in the "page.trend.test" code. The line
that computes "p.table" should have >= instead of > in the
ifelse
statements. E.g., with 3 treatments and 6 blocks the value L=79
has a p-value of 0.0288 (based on Table A.23 in Hollander and
Wolfe), but the function reports "NS" (think of the crushing
disappointment the researcher must have felt :-)
I've attached the code for the example where I discovered this.
The data are from problem 7 in Chapter 4 of the Higgins text
"Introduction to Modern Nonparametric Statistics".
Regards, Rob Kushler
k <- 3
b <- 6
prob47 <- data.frame(day=factor(rep(1:k,each=b)),blk=factor(rep(1:b,k)),
dryhay <-
c(1.5,2.1,1.9,2.8,1.4,1.8,1.8,2.0,2.0,2.7,1.6,2.3,1.9,2.5,2.5,2.6,2.1,2.4))
attach(prob47)
(ymat <- matrix(dryhay,nrow=b,ncol=k))
library(concord)
page.trend.test(ymat)
muL <- b*k*(k+1)2/4
varL <- b*(k2)*(k+1)*(k2 -1)/144
(zL <- (79 - muL)/sqrt(varL))
pnorm(zL,lower=F)
