[R] linear correlation?
David O. Nelson
daven at llnl.gov
Thu Mar 7 19:11:06 CET 2002
I keep on my bulletin board, in plain view of clients at all times, a paper
by David Brillinger with the title "Does Anyone Know When the Correlation
Coefficient is Useful? A Study of the Times of Extreme River Flows"
(Technometrics, 2001). In it he quotes Tukey as follows:
"I frequently hold to the position that correlation coefficients are
justified in two and only two circumstances, when they are regression
coefficients, or when the measurement of one or both variables on a
determinate scale is hopeless."
Neither seems to apply here.
David O Nelson, Ph.D. (daven at llnl.gov)
Lawrence Livermore National Laboratory
Box 808, L-441
Livermore CA 94551
> -----Original Message-----
> From: owner-r-help at stat.math.ethz.ch
> [mailto:owner-r-help at stat.math.ethz.ch]On Behalf Of
> Setzer.Woodrow at epamail.epa.gov
> Sent: Thursday, March 07, 2002 8:02 AM
> To: dechao wang
> Cc: 'R-help at lists.R-project.org'
> Subject: Re: [R] linear correlation?
> To follow up, consider the vector x3 <- c(3,2,1,108, 209, 303) with the
> same units as before.
> > cor(x1,x3)
>  0.9995864
> Now express the first three values as microns instead of cm:
> x4 <- x3
> x4[1:3] <- 10000 * x4[1:3]
> > cor(x1,x4)
>  -0.7461934
> Just changing the units changes the whole sense of the correlation.
> R. Woodrow Setzer, Jr. Phone:
> (919) 541-0128
> Experimental Toxicology Division Fax: (919)
> Pharmacokinetics Branch
> NHEERL MD-74; US EPA; RTP, NC 27711
r-help 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-help-request at stat.math.ethz.ch
More information about the R-help