The apply functions are really just hidden loops, and loops have been made efficient enough that they are usually not much slower (and sometimes a bit faster) than the apply's.
If you really want to use apply, then look at mapply (might need to convert the matrix to a list), or you could use sapply on the vector 1:500 and write a function that indexes into the matrix and vector. But if you understand the loop, then I would suggest using the loop. -- Gregory (Greg) L. Snow Ph.D. Statistical Data Center Intermountain Healthcare greg.s...@imail.org 801.408.8111 > -----Original Message----- > From: r-help-boun...@r-project.org [mailto:r-help-bounces@r- > project.org] On Behalf Of Patrick Boily > Sent: Wednesday, February 02, 2011 1:03 PM > To: 'r-help@r-project.org' > Subject: [R] A question about sampling > > Greetings, > > I am attempting to do something with R that I think should be > efficiently do-able, but I haven't yet found success. > > I have a vector of probability weights (for 17 categories), let's call > it things (it could look like the one below, for instance). > > > things > 0.026 0 0.233 0 0.131 0 0.415 0 0 0 0 0 0.192 0 0.067 0 0 > > I'd like a sample of size size.things (say, 47) of the 17 categories > (with replacement). And I'd like to produce a vector of length 17 which > enumerates the number of times each category has been selected. This is > fairly straightforward to do; for instance: > > > things2<- > table(factor(sample(1:17,size.things[1],replace=TRUE,prob=things),level > s=1:17)) > 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 > 1 0 9 0 4 0 18 0 0 0 0 0 5 0 4 0 0 > > What would I need to do if I had a matrix things (50000 x 17) of > probability weight vectors and a vector of sample sizes size.things (of > length 50000), and I wanted to simultaneously sample size.things[1] of > the 17 categories with probability weight vector things[1,], > size.things[2] of the 17 categories with probability weight vector > things[2,], etc. A loop will do the trick, but it takes a while and it > seems to me that I could more efficiently use tapply somehow. Or > something that behaves like rowSums. I'm not familiar enough with R to > see an easy way out. Perhaps there isn't? Does anybody have an idea? > > Regards, > > Patrick > > > > > > > > > [[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. ______________________________________________ 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.
