[R] Sampling distribution (PDF & CDF) of correlation

From: Mike Lawrence <Mike.Lawrence_at_dal.ca>
Date: Thu, 17 Jul 2008 12:29:20 -0300

Hi all,

I'm looking for an analytic method to obtain the PDF & CDF of the sampling distribution of a given correlation (rho) at a given sample size (N).

I've attached code describing a monte carlo method of achieving this, and while it is relatively fast, an analytic solution would obviously be optimal.

get.cors <- function(i, x, y, N){

	.Internal(cor(x[(end-N+1):end] ,y[(end-N+1):end] ,TRUE ,FALSE ))


get.r.dist <- function(N, rho, it){
	eS = eigen(Sigma, symmetric = TRUE, EISPACK = TRUE)
	ev = eS$values
	fact = eS$vectors %*% diag(sqrt(pmax(ev, 0)), 2)
	Z = rnorm(2 * N * it)
	dim(Z) = c(2, N * it)
	Z = t(fact %*% Z)
	x = Z[, 1]
	y = Z[, 2]
	r = sapply(1:it ,get.cors,x, y, N)

#Run 1e3 monte carlo iterations, where each obtains the correlation
# of 10 pairs of observations from a bivariate normal distribution with
# a true correlation of .5. Returns 1e3 values for the observed  
mc.rs = get.r.dist( N=10 , rho=.5 , it=1e3 )

#plot the PDF & CDF
hist(mc.rs,prob=T,xlab='Observed correlation') probs = seq(0,1,.01)
plot(quantile(mc.rs,probs=probs),probs,type='l',xlab='Observed correlation',ylab='Cumulative probability')

Mike Lawrence
Graduate Student, Department of Psychology, Dalhousie University


"The road to wisdom? Well, it's plain and simple to express:
Err and err and err again, but less and less and less."
	- Piet Hein

R-help_at_r-project.org mailing list
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Received on Thu 17 Jul 2008 - 15:38:08 GMT

Archive maintained by Robert King, hosted by the discipline of statistics at the University of Newcastle, Australia.
Archive generated by hypermail 2.2.0, at Thu 17 Jul 2008 - 21:32:45 GMT.

Mailing list information is available at https://stat.ethz.ch/mailman/listinfo/r-help. Please read the posting guide before posting to the list.

list of date sections of archive