I really like git. It’s the first versioning tool I’ve ever used so I have nothing else to compare it to, but in the world of statistical model building where iteration is constant (and almost never a strict linear progression)...

I really like git. It’s the first versioning tool I’ve ever used so I have nothing else to compare it to, but in the world of statistical model building where iteration is constant (and almost never a strict linear progression)...

After using github for data mining competitions and a project on statistical language models I found I enjoyed it some much I wanted to use it at work too. The trick is there’s a lot of overlap between what I...

The Objective I wanted to source R scripts hosted on my github repository for use in my blog (i.e. a github version of ?source). This would make it easier for anyone wishing to test out my code snippets on their own computers without having to manually go to my github repo and retrieve a series of R

If you dig around enough on Amazon.com, you can find some pretty odd products (like the Badonkadonk tank now sadly unavailable). Attached to these products you can often find a new form of comedy: the funny Amazon review. The products that attract such attention can be hard to fathom: this gallon of milk has more than 1,000 reviews. (Sample:...

Benford's law, also called the first-digit law, states that in lists of numbers from many (but not all) real-life sources of data, the leading digit is distributed in a specific, non-uniform way. According to this law, the first digit is 1 about 30% of the time, and larger digits occur as the leading digit with lower and lower frequency,...

When predicting 0/1 data we can use logit (or probit or robit or some other robust model such as invlogit (0.01 + 0.98*X*beta)). Logit is simple enough and we can use bayesglm to regularize and avoid the problem of separation. What if there are more than 2 categories? If they’re ordered (1, 2, 3, etc),