In an effort to test a theory that a specific author's identity can be determined from a writing sample of sufficient length, I have begun by gathering statistics from a chapter taken from one of my own books. These statistics include number of total words, number of different words, number of words of different length from 1 to 16, letter frequency, bigram frequency, trigram frequency, and some comparisons from published sources.
"Identifying Authors by Lexicostatistics: 1,"
Word Ways: Vol. 33
, Article 10.
Available at: http://digitalcommons.butler.edu/wordways/vol33/iss1/10