Hi Don, Bill and A.K. Thanks for your reply. It worked!
On Fri, Jun 20, 2014 at 3:56 PM, arun <smartpink...@yahoo.com> wrote: > You could try: > library(plyr) > res <- ddply(TestData[,-2],.(ID),numcolwise(max)) > colnames(res)[-1] <- paste0(colnames(res)[-1],".max") > A.K. > > > > > On Friday, June 20, 2014 3:43 PM, Jun Shen <jun.shen...@gmail.com> wrote: > Hi all, > > Here is my situation. I have a dataframe, the structure would be something > like this, > > > TestData<-data.frame(ID=rep(1:10,each=10),TIME=rep(seq(0.1,1,0.1),10),VAR1=rnorm(100),VAR2=5*rnorm(100),VAR3=10*rnorm(100)) > > Basically, I want to extract the maximum value from each ID for VAR1, VAR2, > VAR3...... > > The way I can think of is > > > do.call(rbind,lapply(split(TestData,TestData$ID),function(x)x[which.max(x$VAR1),'VAR1'])) > > and do this for each of the variables and put the results back. It's kind > of clumsy but OK for several variables. I have dozens of them. Is there a > better way to do it? > > It would be ideal to produce the results like > > ID VAR1.max VAR2.max VAR3.max 1 1.2828796 8.63276 15.051992 2 > 1.1870067 > 8.691801 10.736301 3 1.2815352 6.335692 5.827524 4 1.6719411 5.998597 > 16.646212 5 1.5631107 6.067457 15.331046 6 0.718989 6.610279 7.306005 7 > 0.8734315 13.39844 16.965365 8 2.7447862 10.21613 22.545131 9 3.490395 > 10.83543 25.744662 10 0.4719087 11.73021 7.226687 > Thanks for any help. > > Jun Shen > > [[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]] ______________________________________________ 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.
