Thanks all. Dennis provided the answer I was looking for - deviance(lm). Cheers,
Brian On Tue, Jan 25, 2011 at 8:48 PM, Dennis Murphy <djmu...@gmail.com> wrote: > Hi: > > On Tue, Jan 25, 2011 at 10:36 AM, Greg Snow <greg.s...@imail.org> wrote: > >> It is not clear what you are doing or why you are doing it. If you tell >> us your ultimate goal we may be able to help you find a way that does not >> require all the computing that you are doing. >> >> How do you get your coefficients? Are you using lm? Have you looked at the >> resid function? >> > > Thanks, Greg. I totally agree that a clearer explanation of the goals of > this computing problem would be (very) useful. R has a number of ways to > process subsets of data efficiently. One approach (although not unique) > would be to (a) split the data into multiple subgroups; (b) run a lm() model > in each subgroup; (c) export the model objects to a list; (d) extract the > pieces of the output you want for all models run. Several examples of how to > do this with enhanced apply-like functions in the plyr package are in the > list archives, so this can be done with a bit of thinking and an > understanding of what is wanted/needed...especially if you will need to do > multiple sets of extractions from your set of model objects. > > If you find yourself doing the same thing over and over in R, you can bet > there's a much more efficient way to do the computations. An explanation of > the goals, a subset of data that illustrates the problem (preferably in > dput() form) and showing what you've tried would be useful. > > Dennis > >> >> -- >> >> 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 Brian J Mingus >> > Sent: Tuesday, January 25, 2011 9:08 AM >> > To: R-help@r-project.org >> > Subject: [R] Extracting SSE from lm >> > >> > Apologies for this simple question - >> > >> > Given the number of comparisons I need to do it has become somewhat >> > laborious to compute the SSE manually. I first have to extract the >> > coefficients, build the model and run the model on the data. So far I >> > haven't found any method in R that will do this for me. Is there a >> > method >> > that I haven't seen, or is there a small function I could write that >> > would >> > do this, and how might I go about that? >> > >> > Thanks, >> > >> > Brian >> > >> > [[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. >> > > [[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.