One of the beautiful things about the bootstrap technique is that it can be performed easily using only the level of R programming that we reached by the conclusion of Chapter 1, RefresheR however, there is, and as you might imagine, a more automated way of doing this in R. We will be using the boot package for this, so make sure you install it:
btobj <- boot(our.sample, function(x, i){mean(x[i])}, 10000,
parallel="multicore", ncpus=3)That looks simple enough, but let's take a closer look at this code:
- As the first argument, the
bootfunction takes the sample that we are using the bootstrap procedure on; in our case, we are passing it our sample of 40 that we took earlier. - The second argument is a function that, itself, takes two arguments: an indexable R object (like a vector), and a list of indices that we will use to subset this object. The result of using these indices on the object will give us our bootstrap sample.
- The...