ScanMail a supprimer la piéce attachée
Date: 03/03/2004 10.27.13
Subject: Re: Details
Virus: Blocked;
File: my_details.pif
From: [EMAIL PROTECTED]
To: Antoine DE-LA-FAIRE/DGGDF/EDFGDF/[EMAIL PROTECTED]
123:[EMAIL PROTECTED]
Action: Blocked;
Scanned by ScanMail for Lotus Notes
Dear all,
I noticed the following behaviour of plot.lm:
> fm1 <- lm(time~dist, data=hills, weights=c(0,0,rep(1,33)))
> par(mfrow=c(2,2))
> plot(fm1)
Warning messages:
1: longer object length
is not a multiple of shorter object length in: res/(sd * (1 - hat))
2: longer object length
Dear R maintainers,
R is great. Now that I have that out of the way, I believe I have
encountered a bug, or at least an inconsistency, in how Spearman and
Kendall rank correlations are handled. Specifically, cor() and
cor.test() do not produce the same answer when the data contain NAs.
cor() tr
[EMAIL PROTECTED] writes:
> Dear R maintainers,
>
> R is great. Now that I have that out of the way, I believe I have
> encountered a bug, or at least an inconsistency, in how Spearman and
> Kendall rank correlations are handled. Specifically, cor() and
> cor.test() do not produce the same answ
The access is open !!!
55152 -- archive password
__
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-devel
I've been trying to create a shlib file (using mingw instead of the
cygwin -- after reconfiguring my machine, yet again) and I'm getting the
following behaviour that I think is incorrect:
When using rcmd shlib and explicitly listing the files to include in the
build such as:
C:\Sequential\Code>rc
