[R] Jitter in correlation matrix?

From: Mehmet Atif Ergun <mehmetaergun_at_gmail.com>
Date: Sun, 24 Feb 2008 18:26:27 -0500


Hi,

I am just starting to use R for a graduate course, and I like how the correlation matrix at
http://addictedtor.free.fr/graphiques/RGraphGallery.php?graph=137

I did something similar by copying from the examples(pairs), but it seems that I need to jitter the bottom panel... and I have no idea how to do that, and I mean no idea at all. I'd appreciate any help...

Here are the graphs:
http://socy602.pbwiki.com/f/dep_correlationmatrix.jpeg http://socy602.pbwiki.com/f/indep_correlationmatrix.jpeg

And this is the code I used to produce the graphs (not that I understand it, but):

######################

# Primary Component Analyses
######################

######################

# Dependent variables
######################

# List of these variables:
# langfamr , ethdifxx , catness , gc7 , gc8r , gc12 , culdifxx , poldifxx , ecdifxx

######################

# Look at correlation matrix for these

cor(MAR[,c("langfamr" , "ethdifxx" , "catness" , "gc7" , "gc8r" , "gc12" , "culdifxx" , "poldifxx" , "ecdifxx" )], use="complete.obs")
# Big time correlation matrix

 panel.cor <- function(x, y, digits=2, prefix="", cex.cor) {

    usr <- par("usr"); on.exit(par(usr))     par(usr = c(0, 1, 0, 1))
    r <- abs(cor(x, y))
    txt <- format(c(r, 0.123456789), digits=digits)[1]     txt <- paste(prefix, txt, sep="")
    if(missing(cex.cor)) cex <- 0.8/strwidth(txt)     text(0.5, 0.5, txt, cex = cex * r)
}
MAR.omitdep <- na.omit(MAR[,c("langfamr" , "ethdifxx" , "catness" , "gc7" , "gc8r" , "gc12" , "culdifxx" , "poldifxx" , "ecdifxx" )]) pairs(MAR.omitdep[,c("langfamr" , "ethdifxx" , "catness" , "gc7" , "gc8r" , "gc12" , "culdifxx" , "poldifxx" , "ecdifxx" )], lower.panel=panel.smooth, upper.panel=panel.cor)
# Save it

dev.copy(jpeg,filename="dep_correlationmatrix.jpeg",height=600, width=800,bg="white") dev.off()

######################

# Independent variables
######################

# List of variables
# poldis , polres , ecdis , culres , gcc1

######################

# Look at the correlation matrix

cor(MAR[,c("poldis" , "polres" , "ecdis" , "culres" , "gcc1" )], use="complete.obs")
# Big time correlation matrix

 panel.cor <- function(x, y, digits=2, prefix="", cex.cor) {

    usr <- par("usr"); on.exit(par(usr))     par(usr = c(0, 1, 0, 1))
    r <- abs(cor(x, y))
    txt <- format(c(r, 0.123456789), digits=digits)[1]     txt <- paste(prefix, txt, sep="")
    if(missing(cex.cor)) cex <- 0.8/strwidth(txt)     text(0.5, 0.5, txt, cex = cex * r)
}
MAR.omitindep <- na.omit(MAR[,c("poldis" , "polres" , "ecdis" , "culres" , "gcc1" )]) pairs(MAR.omitindep[,c("poldis" , "polres" , "ecdis" , "culres" , "gcc1" )], lower.panel=panel.smooth, upper.panel=panel.cor)
# Save it

dev.copy(jpeg,filename="indep_correlationmatrix.jpeg",height=600, width=800,bg="white") dev.off()

Thanks a lot in advance,
Sincerely,
Mehmet.



R-help_at_r-project.org 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. Received on Sun 24 Feb 2008 - 23:29:34 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 Mon 25 Feb 2008 - 08:30:16 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