Bayesian Analysis of a Multinomial Sequence and Homogeneity of Literary Style

Abstract

To help settle the debate around the authorship of Tirant lo Blanc, all words in each chapter are categorized according to their length, and the appearances of certain words are counted, thus forming two contingency tables of ordered rows. A Bayesian multinomial change-point analysis of the sequence of rows, reveals a clear stylistic boundary, estimated to be near chapters 371 and 382. A Bayesian cluster analysis of these rows confirms the existence of that boundary, and reveals a few chapters that are misclassified by the estimated change-point. The statistical evidence supports the hypotheses of one main author writing about four fifths of the book, with a second author finishing the book by filling in material, mainly at the end of it.