[R] Problem labelling arrays

downunder03 larsfromspace at web.de
Tue Jan 2 23:50:30 CET 2007


Hi all. I'am sitting for hours. I have a problem labelling arrays. I am
loosing labels for higher dimensions than matrices. Has anyone already faced
that problem? There must be some direct way without writing complicated
functions. Thanks for any hint. greetings lars


x<-matrix(sin(1:90),nrow=100,ncol=9)
dimnames(x)[[2]]<-c("F1_5", "F2_5", "F3_5","F4_5", "F5_5","F6_5", "F7_5",
"F8_5", "F9_5")

partialCorr <- function (cond, mat)
        cor(qr.resid(qr(cbind(1, mat[, cond])), mat[, -cond])) 

# matrix keep label
d=partialCorr(c(3,2),x)
d

# array looses label
bsp1 <- array(0, c(7,7,36))
k <- 0
for(i in 1:8) for(j in (i+1):9) {
    k <- k+1
    bsp1[, , k] <- partialCorr(c(i, j), x)
} 
bsp1
# I tried something to label inside the loop
bsp1 <- array(0, c(7,7,36))
k <- 0
for(i in 1:8) for(j in (i+1):9) {
    k <- k+1
    bsp1[, , k] <- partialCorr(c(i, j), x)
    dimnames(bsp1[,,k])=dimnames(partialCorr(c(i,j), x))
} 
bsp1
# I tried something to label outside the loop but can't directy adress the
array
labels <- array(0, c(7,1,36))
k <- 0
for(i in 1:8) for(j in (i+1):9) {
    k <- k+1
    labels[, , k] <- as.vector(rownames(partialCorr(c(i,j), x)))
}
labels
for (k in 1:36){
dimnames(bsp1[,,k])=list(c(labels[,1,k]),c(labels[,1,k]))
}
bsp1

-- 
View this message in context: http://www.nabble.com/-R--Problem-labelling-arrays-tf2910585.html#a8132366
Sent from the R help mailing list archive at Nabble.com.



More information about the R-help mailing list