Publications by John Myles White
Forecasting Presidential Elections
Because of Andrew Gelman’s strong, repeated recommendations, I’ve been reading “Forecasting Presidential Elections” by Steven J. Rosenstone for the last two days. It’s quite a remarkable book and complex enough that I’m sure I’ll return to it many times after I’ve finished it. I was particularly intrigued by a table in the first c...
2601 sym 2 img
Visualizing Eigenfactors
These interactive graphics are simply beautiful. And they just so happen to be profoundly informative about the structure of modern science as well. Here’s to the hope that we will see more work from Moritz Stefaner soon that shows how our aesthetic and scientific demands can be met simultaneously. HT to Infosthetics. Related To leave a comme...
746 sym
If I Had a Text File, I’d Hack Regexes in the Morning
Yesterday the topic of academic citation counts came up, so I decided that I should write up some tools for exploring cite counts. The first thing I did was to build a cheap screenscraper in Ruby for pulling citation count information from Google scholar. You’ll see the ugly hack I produced below. 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19...
1183 sym Python (2244 sym/4 pcs) 2 img 2 tbl
Princeton Graduate Student Housing
For any Princeton graduate students who are interested, here’s the success rate for graduate students applying for school housing. These charts were built using the data from the 2008-2009 Room Draw Statistics pamphlet provided by the Division of Housing here at Princeton. Related To leave a comment for the author, please follow the link and ...
705 sym 10 img
Single Letter Frequencies in English
Every time that I read a paper that discusses the frequencies of single letters in English, I feel like I should sit down and calculate them for myself from a sample of English text. Today, I finally did. Here are the probabilities and negative log probabilities of the characters in English over the corpus of Shakespeare’s plays: And, for thos...
882 sym R (898 sym/2 pcs) 4 img 1 tbl
Pearson vs. Spearman Correlation Coefficients
One of the misuses of statistical terminology that annoys me most is the use of the word “correlation” to describe any variable that increases as another variable increases. This monotonic trend seems worth looking for, but it plainly is not what most people discover when they use standard correlation coefficients. This is because the Pearson...
1800 sym 2 img
Color Schemes for R Bar Plots
A recurrent source of irritation for me is the absence of a good default behavior in R for choosing the color scheme for bar plots. A stacked bar plot looks only as good as the color scheme you use. In hope of finding a usable scheme that I could settle on as a personal default, I picked two color schemes, Sunshine over Glacier and Sweet Valentin...
1404 sym 4 img
Click Tracks and Beat Detection
Being a drummer, a programmer and a fan of statistical analysis, this post on the (unnaturally) perfect timing of drum parts recorded to a click track was a real delight to me. Of course, many claims in the post are odd: it seems hard to imagine that a person recorded the drums for Britney Spears. And it seems as likely that some of the drum par...
1037 sym
Wanderlust
We Americans have a reputation as being unworldly. Given the results of the most recent Pew survey, perhaps we deserve it. Evidently, the majority of us never move out of our home states. Related To leave a comment for the author, please follow the link and comment on their blog: "R-bloggers" via Tal Galili in Google Reader. R-bloggers.com of...
614 sym
Causation’s Mistreated Sibling Correlation
This is why I love XKCD, though surely the best part of this strip was the mouseover: “correlation doesn’t imply causation, but it does waggle its eyebrows suggestively and gesture furtively while mouthing, ‘look over there’.” Related To leave a comment for the author, please follow the link and comment on their blog: "R-bloggers" vi...
654 sym 2 img