Here's an amusing story. I can understand why the guy computed his test from scratch, but I agree with Dan that the two-page appendix is kind of over the top.
Deriving significance tests from scratch
2 Comments
Leave a comment
Subscribe to Entry
Email:
Subscribe!
Andrew Gelman's Books
I once discovered that I had rederived Kendall's tau. I was actually pretty proud, and it meant we could replace a page of justification of this new test we invented with "we used Kendalls' tau" (luckily we caught it before the final version -- of course, it had been to several reviewers who didn't notice ...)
Sometimes "proper tools" are too general. And your code for your case can be much more efficient.