Generating all subsets of a set

[This article was first published on R snippets, and kindly contributed to R-bloggers]. (You can report issue about the content on this page here)
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Recently I have calculated Banzhaf power index. I required generation of all subsets of a given set. The code given there was a bit complex and I have decided to write a simple function calculating it. As an example of its application I reproduce Figure 3.5 from Hastie et al. (2009).
The figure shows RSS for all possible linear regressions for prostate cancer data on training subset. The standard approach for such a problem in R is to use leaps package, but I simply wanted to test my function generating all subsets of the set.

Here is the code with all.subsets function generating all subsets and its application to prostate cancer data:


all.subsets <- function(set) {
    n <- length(set)
    bin <- expand.grid(rlply(n, c(F, T)))
    mlply(bin, function() { set[c()] })

file.url <- “”
data.set <- read.table(file.url, sep = “\t”, header = TRUE)
varlist <- all.subsets(names(data.set)[2:9])

get.reg <- function(vars) {
    if (length(vars) == 0) {
        vars = “1”
    vars.form <- paste(“lpsa ~”, paste(vars, collapse = ” + “))
    lm(vars.form, data = data.set, subset = train)

models <- llply(varlist, get.reg)
models.RSS <- ldply(models, function(x) {
    c(RSS = sum(x$residuals ^ 2), k = length(x$coeff)) })

min.models <- ddply(models.RSS, .(k), function(x) {
qplot(k, RSS, data = models.RSS, ylim= c(0,100),
    xlab = “Subset Size k”, ylab = “ResidualSum-of-Squares”,) +
geom_point(data = min.models, aes(x = k, y = RSS), colour = “red”) +
geom_line(data = min.models, aes(x = k, y = RSS), colour = “red”) +

And here is the plot it generates:

To leave a comment for the author, please follow the link and comment on their blog: R snippets. offers daily e-mail updates about R news and tutorials about learning R and many other topics. Click here if you're looking to post or find an R/data-science job.
Want to share your content on R-bloggers? click here if you have a blog, or here if you don't.

Never miss an update!
Subscribe to R-bloggers to receive
e-mails with the latest R posts.
(You will not see this message again.)

Click here to close (This popup will not appear again)