subgroup-based quantiles

Hi,
You could also try ?ave()
dat$Age_group<-ave(dat$Age,dat$GENDER,FUN=function(x){cut(x,labels=1:10,breaks=quantile(x,seq(0,1,.1)),include.lowest=TRUE)})
  dat
#  GENDER Age Age_group
#1      2  45         1
#2      1  58        10
#3      1  54         1
#4      2  71        10
#5      2  64         5
A.K.



----- Original Message -----
From: Rui Barradas <ruipbarradas at sapo.pt>
To: R Kozarski <r.kozarski at gmail.com>
Cc: r-help at r-project.org
Sent: Friday, November 30, 2012 1:07 PM
Subject: Re: [R] subgroup-based quantiles

Hello,

If you want Age quantiles by gender, you have to split the data by
gender, apply the same code then recombine the result.


fun <- function(x){
      Age_group <- cut(x[, "Age"], labels=c(1:10),
          breaks=quantile(x[, "Age"], seq(0,1,.1)),
          include.lowest = TRUE)
      cbind(x, Age_group)
}

result <- do.call(rbind, lapply(split(dat, dat[, "GENDER"]), fun))
rownames(result) <- seq_len(nrow(result))
result


Hope this helps,

Rui Barradas
Em 30-11-2012 12:18, R Kozarski escreveu:

Dear R users,
given the patient sample with their Gender and Age
           GENDER Age
      [1,]      2  45
      [2,]      1  58
      [3,]      1  54
      [4,]      2  71
      [5,]      2  64
    ...
I would like to create an another column which groups the patients wrt
Gender specific Age quantiles, following methodology similar to:

Age_group <- cut(Age, labels=c(1:10), breaks=quantile(Age,
seq(0,1,.1)),include.lowest = TRUE)

The function above allows me to group only wrt Age quantiles.

Best, Robert




--
View this message in context: http://r.789695.n4.nabble.com/subgroup-based-quantiles-tp4651415.html
Sent from the R help mailing list archive at Nabble.com.