Sorry about the formatting in that previous email. Here it is
again, hopefully with the correct formatting this time.
Assuming that the data frame is called m, you can
- split the data frame by the values of siteid and date
- apply a function to create the new data frame for each such group and
- use rbind to put it back together like this:
fn <- function(x) {
y <- t(x[,4:5])
data.frame( siteid=x[1,1], date=x[1,2], name=colnames(x)[4:5], x01=y[,1], x02=y[,2],
x03=y[,3] )
}
m <- do.call( "rbind", lapply(split(m,list(m$siteid,m$date)),fn) )
If order matters, add this:
m <- m[order(m$siteid,m$date),]
--- On Mon 09/29, < [EMAIL PROTECTED] > wrote:
From: [mailto: [EMAIL PROTECTED]
To: [EMAIL PROTECTED]
Date: Mon, 29 Sep 2003 13:15:36 -0400
Subject: [R] Data frame transpose
<br><br><br><br>Hi All,<br><br>I want to ask if there is a transpose function for data
frame like the<br>procedure of transpose in SAS? Because I want to partially transpose
a<br>data frame which contains 5 columns (siteid, date, time, obs, mod), what<br>I
want to do is to put time as the column variables along with siteid,<br>and date, and
put obs and mod in the row names. specifically to<br>transpose a data
frame:<br><br>siteid date time obs mod<br> A 7/8 01
2 5<br> A 7/8 02 3 8<br> A 7/8
03 5 8<br> A 7/9 01 3 6<br> A
7/9 02 5 8<br> A 7/9 03 6 7<br>
......<br> B 7/8 01 4 7<br> B
7/8 02 7 19<br> B 7/8 03 4 9<br>
......<br><br>To<br><br>siteid date!
name 01 02 03 ....<br>A 7/8 obs 2 3
5<br>A 7/8 mod 5 8 8<br>A 7/9
obs 3 5 6<br>A 7/9 mod 6 8
7<br> .......<br>B 7/8 obs 4 7
4<br>B 7/8 mod 7 19 9<br>
........<br><br><br>Thank you very
much!<br><br>Dave<br><br>______________________________________________<br>[EMAIL
PROTECTED] mailing list<br>https://www.stat.math.ethz.ch/mailman/listinfo/r-help<br>
_______________________________________________
No banners. No pop-ups. No kidding.
Introducing My Way - http://www.myway.com
______________________________________________
[EMAIL PROTECTED] mailing list
https://www.stat.math.ethz.ch/mailman/listinfo/r-help
