newdats <- rbind(cbind(dats[rep(1:nrow(dats), dats$AEs), 1:2], AEs=1), cbind(dats[rep(1:nrow(dats), dats$N-dats$AEs),1:2], AEs=0))
But the data will not be in the order you specified unless you add newdats <- newdats[order(newdats$Study, -newdats$TX, -newdats$AEs),] and you may want to clean up the rownumbers with rownames(newdats) <- 1:nrow(newdats) ---------------------------------------------- David L Carlson Associate Professor of Anthropology Texas A&M University College Station, TX 77843-4352 > -----Original Message----- > From: r-help-boun...@r-project.org [mailto:r-help-bounces@r- > project.org] On Behalf Of R. Michael Weylandt > Sent: Tuesday, May 15, 2012 1:09 AM > To: Cheenghee AM Koh > Cc: r-help@r-project.org > Subject: Re: [R] How to Un-group a grouped data set? > > It is a nifty and surprisingly useful construct whenever you need to > construct a function call programmatically or apply it to a list. > > R-News 2/2 has some useful tips on this and related functions in the > Programmer's Note section if you're interested. > > Best, > Michael > > On Tue, May 15, 2012 at 2:05 AM, Cheenghee AM Koh <sigo...@gmail.com> > wrote: > > Thank you so much! I can't believe I spent the whole night by not > knowing > > this one command "do.call" > > This is so handy! > > Best, Koh > > > > > > On Tue, May 15, 2012 at 12:52 AM, R. Michael Weylandt > > <michael.weyla...@gmail.com> wrote: > >> > >> Sorry -- I missed the bit about the AE in your original post. > Perhaps > >> you can work with my bit for the repeats, but it looks like if you > >> want to use your function, it should suffice to do something like > >> > >> do.call("rbind", lapply(NewFuncName, 1:6)) > >> > >> Best, > >> Michael > >> > >> On Tue, May 15, 2012 at 1:50 AM, R. Michael Weylandt > >> <michael.weyla...@gmail.com> wrote: > >> > Don't use subset for a function name -- it's already the name of a > >> > rather important function as is data (but at least that one's not > a > >> > function in your use so it's not quite so bad). Finally, use > dput() > >> > when sending data so we get a plaintext reproducible version. > >> > > >> > I'd try something like this: > >> > > >> > dats <- structure(list(Study = c(1L, 1L, 2L, 2L, 3L, 3L), TX = > c(1L, > >> > 0L, 1L, 0L, 1L, 0L), AEs = c(3L, 2L, 1L, 2L, 1L, 1L), N = c(5L, > >> > 7L, 10L, 7L, 8L, 4L)), .Names = c("Study", "TX", "AEs", "N"), > class = > >> > "data.frame", row.names = c("1", > >> > "2", "3", "4", "5", "6")) > >> > > >> > # See how handy dput can be :-) > >> > > >> > dats[unlist(mapply(FUN = function(x,y) rep(x, y), 1:NROW(dats), > >> > dats$N)), -4] > >> > > >> > which isn't super elegant, but others might have something better. > >> > > >> > Best, > >> > Michael > >> > > >> > On Tue, May 15, 2012 at 1:24 AM, Cheenghee AM Koh > <sigo...@gmail.com> > >> > wrote: > >> >> Hello, R-fellows, > >> >> > >> >> I have a question that I really don't know how to solve. I have > spent > >> >> hours > >> >> on line surfing for possible solutions but in veil. Please if > anyone > >> >> could > >> >> help me handle this issue, you would be so appreciated! > >> >> > >> >> I have a "grouped" dataset like this: > >> >> > >> >>> data > >> >> Study TX AEs N > >> >> 1 1 1 3 5 > >> >> 2 1 0 2 7 > >> >> 3 2 1 1 10 > >> >> 4 2 0 2 7 > >> >> 5 3 1 1 8 > >> >> 6 3 0 1 4 > >> >> > >> >> where Study is the study id, TX is treatment, AEs is how many > people in > >> >> this trial is positive, and N is the number of the subjects. > Therefore, > >> >> for > >> >> the row 1, it stands for: It is the treatment arm for the study > one, > >> >> where > >> >> there are 5 subjects and 3 of them are positive. The row 2 stands > for: > >> >> It > >> >> is the control arm of the study 1 where there are 7 subjects and > 2 of > >> >> them > >> >> are positive. > >> >> > >> >> Now I would like to "un-group them", make it like: > >> >> > >> >> Study TX AEs > >> >> 1 1 1 > >> >> 1 1 1 > >> >> 1 1 1 > >> >> 1 1 0 > >> >> 1 1 0 > >> >> 1 0 1 > >> >> 1 0 1 > >> >> 1 0 0 > >> >> 1 0 0 > >> >> 1 0 0 > >> >> 1 0 0 > >> >> 1 0 0 > >> >> 2 1 1 > >> >> ..................... > >> >> ..................... > >> >> > >> >> > >> >> But I wasn't able to do it. In fact I wrote a small function, and > use > >> >> "lapply" to get what I want. It worked well, and did give me what > I > >> >> want. > >> >> But I wasn't able to collapse all the returns into one single > data > >> >> frame > >> >> for subsequent analysis. > >> >> > >> >> The function I wrote: > >> >> > >> >> subset = function(i){ > >> >> d = c(rep(data[i,1], data[i,4]), rep(data[i,2], data[i,4]), > rep(0:1, > >> >> c(data[i,4] - data[i,3],data[i,3]))) > >> >> d = matrix(d, data[i,4],3) > >> >> d > >> >> } > >> >> > >> >> then: > >> >> > >> >> Data = lapply(1:6, subset) > >> >> Data > >> >> > >> >> Therefore, I tried to write a loop. But no matter how I tried, I > can't > >> >> get > >> >> what I want. > >> >> > >> >> Any idea? > >> >> > >> >> Thank you so much! > >> >> > >> >> Best, > >> >> > >> >> > >> >> -- > >> >> Cheenghee Masaki Koh, MSW, MS(c), PhD Student > >> >> School of Social Service Administration > >> >> Department of Health Studies, Division of Biological Science > >> >> University of Chicago > >> >> > >> >> [[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. > > > > > > > > > > -- > > Cheenghee Masaki Koh, MSW, MS(c), PhD Student > > School of Social Service Administration > > Department of Health Studies, Division of Biological Science > > University of Chicago > > > > ______________________________________________ > 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.
