HI, May be this helps: dat1<-read.table(text=" id status week 1 no 1 1 no 2 1 no 3 1 no 4 1 no 5 1 no 6 1 no 7 2 no 1 2 no 2 2 no 3 2 no 4 2 yes 5 2 yes 6 2 na 7 2 na 8 2 na 9 3 no 1 3 no 2 3 no 3 3 Unknown 4 3 unknown 5 3 na 6 3 na 7 3 na 8 ",sep="",header=TRUE,stringsAsFactors=FALSE,na.strings="na") dat2<-dat1[complete.cases(dat1),] res<-do.call(rbind,lapply(split(dat2,dat2$id),function(x) rbind(tail(x[all(x[,2]=="no")],1),head(x[x[,2]=="yes"|x[,2]=="Unknown",],1)))) res # id status week #1 1 no 7 #2 2 yes 5 #3 3 Unknown 4 A.K.
----- Original Message ----- From: Rich Shepard <rshep...@appl-ecosys.com> To: R help <r-help@r-project.org> Cc: Sent: Friday, January 18, 2013 12:18 PM Subject: Re: [R] longitudinal study On Fri, 18 Jan 2013, bibek sharma wrote: > I have a data set from a longitudinal study ( sample below) where subjects > are followed over time. Second column (status) contains info about if > subject is dead or still in the study and third column is time measured in > the week. Here is what I need: if status is not dead or unknown take the > last week, if status is dead or unknown I need to have corresponding week. > > Desired resulst: > > 1 no 7 > 2 yes 5 > 3 Unknown 4 Looks like a survival analysis situation. I know there are R packages for this. Rich ______________________________________________ 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. ______________________________________________ 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.
