R help - Sep 2006 - Covariance/Correlation matrix for repeated measures data frame

All,
 
I have a repeated measures data frame and was wondering if the
covariance matrix can be 
calculated via some created indexing or built-in R function.
 
Specifically, say there are 3 variables, where potassium concentration
is measured 6 times on each patient.
Patient number (discrete)
Time (1 to 6, discrete)
Potassium (continuous variable)
 
I want the covariance/correlation matrix for the cov/corr between
Potassium at time i and time j.
 
Is this possible in the current dataframe format?  Or do I have to
define new varialbes, say Time i and Time j, 
and then compute the cov/corr between Time i and Time j for all
combinations?   
 
Cheers,
Dave
 
 
 
 
David Afshartous, PhD
University of Miami
School of Business
Rm KE-408
Coral Gables, FL 33124
 

	[[alternative HTML version deleted]]

try the following:

dat <- data.frame(id = rep(1:100, each = 6), time = rep(1:6, 100), pot 
= rnorm(600))

# for a balanced data-set
mat <- matrix(dat$pot, ncol = 6, byrow = TRUE)
cor(mat)

# for a unbalanced data-set
dat <- dat[-sample(600, 100), ]
mat <- t(sapply(split(dat, dat$id), function(x){
    out <- rep(NA, 6)
    out[x$time] <- x$pot
    out
}))

cor(mat, use = "pairwise.complete.obs")


I hope it helps.

Best,
Dimitris

----
Dimitris Rizopoulos
Ph.D. Student
Biostatistical Centre
School of Public Health
Catholic University of Leuven

Address: Kapucijnenvoer 35, Leuven, Belgium
Tel: +32/(0)16/336899
Fax: +32/(0)16/337015
Web: http://med.kuleuven.be/biostat/
     http://www.student.kuleuven.be/~m0390867/dimitris.htm


----- Original Message ----- 
From: "Afshartous, David" <afshart at exchange.sba.miami.edu>
To: <r-help at stat.math.ethz.ch>
Sent: Wednesday, September 06, 2006 5:59 PM
Subject: [R] Covariance/Correlation matrix for repeated measures data 
frame

> All,
>
> I have a repeated measures data frame and was wondering if the
> covariance matrix can be
> calculated via some created indexing or built-in R function.
>
> Specifically, say there are 3 variables, where potassium 
> concentration
> is measured 6 times on each patient.
> Patient number (discrete)
> Time (1 to 6, discrete)
> Potassium (continuous variable)
>
> I want the covariance/correlation matrix for the cov/corr between
> Potassium at time i and time j.
>
> Is this possible in the current dataframe format?  Or do I have to
> define new varialbes, say Time i and Time j,
> and then compute the cov/corr between Time i and Time j for all
> combinations?
>
> Cheers,
> Dave
>
>
>
>
> David Afshartous, PhD
> University of Miami
> School of Business
> Rm KE-408
> Coral Gables, FL 33124
>
>
> [[alternative HTML version deleted]]
>
> ______________________________________________
> 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
> and provide commented, minimal, self-contained, reproducible code.
> 

Disclaimer: http://www.kuleuven.be/cwis/email_disclaimer.htm