On 27-04-2014, at 11:01, Berend Hasselman <b...@xs4all.nl> wrote:
>
> On 27-04-2014, at 02:37, Nico Met <nicome...@gmail.com> wrote:
>
>> Dear all,
>>
>> I have a matrix (dimension, 16 x 12) where 2nd column represents class
>> (1,1,1,1,1,2,2,2, etc) information. I want to estimate average and median
>> values for each of the class and add this information as a row at end of
>> the each classes.
>> …...
>>
>> And same will be for other classes.
>
> ……
> For each “class” (group would ba better name) you must append something
> different e.g. the “class”-number.
>
> Try this:
>
> library(plyr)
> g <- function(dat) {
> avg <- as.data.frame(aggregate(dat[,2:dim(dat)[2]], dat["class"],
> function(x) mean(x,na.rm=T)) )
> med <- as.data.frame(aggregate(dat[,2:dim(dat)[2]], dat["class"],
> function(x) median(x,na.rm=T)) )
> z <- rbind(dat,avg,med)
> z
> }
> DAT1 <- ddply(dat,.(class),.fun=g)
> rownames(DAT1) <- do.call(c,lapply(split(dat,dat["class"]),
> FUN=function(x)
> c(rownames(x),paste0("Avg",x[,"class"][1]),paste0("Med",x[,"class"][1]))))
> DAT1
>
> Convoluted but it works. Maybe someone else can come up with something
> shorter and more elegant.
Or something like this if you don’t want to use package plyr
g <- function(dat) {
avg <- as.data.frame(aggregate(dat[,2:dim(dat)[2]], dat["class"],
function(x) mean(x,na.rm=T)) )
med <- as.data.frame(aggregate(dat[,2:dim(dat)[2]], dat["class"],
function(x) median(x,na.rm=T)) )
z <- rbind(dat,avg,med)
rownames(z) <-
c(rownames(dat),paste0("Avg",dat[,"class"][1]),paste0("Med",dat[,"class"][1]))
print(z)
z
}
D <- lapply(split(dat,dat["class"]),FUN=g)
DAT2 <- do.call(rbind, lapply(D, data.frame))
# alternatively
# DAT2 <- do.call(rbind, lapply(D,FUN=function(x) x ))
rownames(DAT2) <- sub("[0-9]+\\.","",rownames(DAT2))
DAT2
Berend
______________________________________________
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.
