R help - Oct 2011 - plots of correlation matrices

Hi,

I want to do a visualisation of a matrix plot made up of several plots of
correlation matrices (using corrplot()). My data is in csv format. Here's an
example:

id,category,attribute1,attribute2,attribute3,attribute4
661,SCHS,43.2,0,56.5,1
12202,SCHS,161.7,5.7,155,16
1182,SCHS,21.4,0,29,0
1356,SSS, 8.8182,0.1818,10.6667,0.6667
1864,SCHS,443.7273,9.9091,537,46
12360,SOA,6.6364,0,10,0
3382,SOA,7.1667,0,26,0.5
1033,SOA,63.9231,1.5385,91.5,11.5
14742,SSS,4.3846,0,8,0
12760,SSS,425.0714,1.7857,297.5,3.5

I can get rid of the id. But I need the 'category' as a way of
distinguishing the various correlation matrices.
I can do a plot of the correlation matrix using corrplot() function in the
corrplot package (ignoring the id and category). But what I need is a matrix
of the plots of each correlation matrix based on the category, ie I have
three categories in the data, hence I will need three plots of the
correlation matrix  in one diagram (because the correlation matrix only
makes sense if they are distinguished by category).

Any help?

Regards
Gawesh

	[[alternative HTML version deleted]]

Hi,

One way to do that is this  (avoiding the use of a for loop):


l.txt<- "id category attribute1 attribute2 attribute3 attribute4
661 SCHS 43.2 0 56.5 1
12202 SCHS 161.7 5.7 155 16
1182 SCHS 21.4 0 29 0
1356 SSS  8.8182 0.1818 10.6667 0.6667
1864 SCHS 443.7273 9.9091 537 46
12360 SOA 6.6364 0 10 0
3382 SOA 7.1667 0 26 0.5
1033 SOA 63.9231 1.5385 91.5 11.5
14742 SSS 4.3846 0 8 0
12760 SSS 425.0714 1.7857 297.5 3.5
"

dat.df <- read.table(textConnection(l.txt),  header=T, as.is = TRUE)
closeAllConnections()

dat.lt<-by(dat.df[,3:6], dat.df$category, cor)
lapply(dat.lt,corrplot)


Regards,
Carlos Ortega
www.qualityexcellence.es

2011/10/11 gj <gawesh@gmail.com>
> Hi,
>
> I want to do a visualisation of a matrix plot made up of several plots of
> correlation matrices (using corrplot()). My data is in csv format.
Here's
> an
> example:
>
> id,category,attribute1,attribute2,attribute3,attribute4
> 661,SCHS,43.2,0,56.5,1
> 12202,SCHS,161.7,5.7,155,16
> 1182,SCHS,21.4,0,29,0
> 1356,SSS, 8.8182,0.1818,10.6667,0.6667
> 1864,SCHS,443.7273,9.9091,537,46
> 12360,SOA,6.6364,0,10,0
> 3382,SOA,7.1667,0,26,0.5
> 1033,SOA,63.9231,1.5385,91.5,11.5
> 14742,SSS,4.3846,0,8,0
> 12760,SSS,425.0714,1.7857,297.5,3.5
>
> I can get rid of the id. But I need the 'category' as a way of
> distinguishing the various correlation matrices.
> I can do a plot of the correlation matrix using corrplot() function in the
> corrplot package (ignoring the id and category). But what I need is a
> matrix
> of the plots of each correlation matrix based on the category, ie I have
> three categories in the data, hence I will need three plots of the
> correlation matrix  in one diagram (because the correlation matrix only
> makes sense if they are distinguished by category).
>
> Any help?
>
> Regards
> Gawesh
>
>        [[alternative HTML version deleted]]
>
> ______________________________________________
> R-help@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.
>
	[[alternative HTML version deleted]]