I have often needed to do this: split a factor variable into a set of dummies. For example, when randomForest will not let you use a factor with more than 32 levels, you may want to take this route. Or if you are like me and don’t like to rely too much on R’s formula interface it may help you choose your base category for a factor or supress intercept easily.

I used to love stackoverflow when it was a read-only site for me. Since I have started participating to answer questions I have been having some mixed feelings about the site. Here are my thoughts in no relevant order.

Since I discovered it, I have found logical subsetting in R a very elegant idiom. Recently I’ve had a change of heart due to two observations. Here is why I propose staying away from it.

If you work in finance, have you ever needed to identify the stocks that were in an index at any given date? Or perhaps a series of dates? Given how central indices are to financial markets and how frequently they change, this is a common problem.

Octopress is a blogging framework for hackers!

This is yet another post from an Octopress starter in praise of the platform. For me, it is an ode to the minimalism and simplicity of this tool that I feel compelled to proclaim the happiness I derive from it.¹