R help - Dec 2002 - splitting columns into groups

Hi,
I'm trying to split 96 columns into 8 groups and then obtain row totals for
each
of those groups. But:

c.factor <- as.factor(c(rep(1:8,c(18,12,12,12,6,12,12,12)))
data.groups <- split(data,c.factor)

splits my rows into those groups. Is there anyway I can tell R to apply the
split function to columns or am I doing something wrong when creating
"c.factor"

Cheers, Jeremy

I think the following does what you want:

data.rowsum <- apply(data, 1, function(x) tapply(x, c.factor, sum))

BTW, I don't think you need to have c.factor as a factor.

HTH,
Andy

-----Original Message-----
From: Jeremy Z Butler [mailto:jeremybutler at paradise.net.nz]
Sent: Wednesday, December 11, 2002 10:06 PM
To: r-help at stat.math.ethz.ch
Subject: [R] splitting columns into groups


Hi,
I'm trying to split 96 columns into 8 groups and then obtain row totals for
each
of those groups. But:

c.factor <- as.factor(c(rep(1:8,c(18,12,12,12,6,12,12,12)))
data.groups <- split(data,c.factor)

splits my rows into those groups. Is there anyway I can tell R to apply the
split function to columns or am I doing something wrong when creating
"c.factor"

Cheers, Jeremy

______________________________________________
R-help at stat.math.ethz.ch mailing list
http://www.stat.math.ethz.ch/mailman/listinfo/r-help

------------------------------------------------------------------------------