Tena koe Philip An alternative to Jim's solution which seems to work and you may, or may not, find less messy:
df <- read.table(text="id yr val a 1950 1 b 1950 10 a 1951 2 b 1952 3 c 1952 4 a 1954 5 b 1954 2 c 1954 3",header=TRUE) df1 <- df[order(df$id, df$yr),] df1$valCS <- unlist(by(df1$val, df1$id, cumsum)) df1 If you need to get back to the original order you can sort by row.names. HTH .... Peter Alspach -----Original Message----- From: r-help-boun...@r-project.org [mailto:r-help-boun...@r-project.org] On Behalf Of Jim Lemon Sent: Monday, 10 March 2014 3:04 p.m. To: Philip A. Viton Cc: r-help@r-project.org Subject: Re: [R] cum sums On 03/10/2014 05:57 AM, Philip A. Viton wrote: > > Suppose I have a dataframe beginning: > > id yr val > a 1950 1 > b 1950 10 > a 1951 2 > > I'm trying to produce a table of cumulative sums of val, disaggregated > by id and then yr, so the result should begin > > id yr cumval > a 1950 1 > a 1951 3 > b 1950 10 > > I've been trying to do this using "aggregate" > and passing the function "cumsum," but I can't get it to work. Can > someone tell me how to do this? Hi Philip, Kinda messy, but try this: df<-read.table(text="id yr val a 1950 1 b 1950 10 a 1951 2 b 1952 3 c 1952 4 a 1954 5 b 1954 2 c 1954 3",header=TRUE) dfc<-by(df[,2:3],df$id,cumsum) ids<-names(dfc) dfcc<-cbind(rep(ids[1],dim(dfc[[1]])[1]),dfc[[1]]) names(dfcc)<-names(df) for(n in 2:length(dfc)) { dfpc<-cbind(rep(ids[n],dim(dfc[[n]])[1]),dfc[[n]]) names(dfpc)<-names(df) dfcc<-rbind(dfcc,dfpc) } Jim ______________________________________________ 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. The contents of this e-mail are confidential and may be ...{{dropped:14}} ______________________________________________ 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.
