Data Analysis Using Regression and Multilevel/Hierarchical Models
Andrew Gelman heard from his publisher:
I regret to inform you that we have been notified by our partner in China, Posts & Telecommunications Press (PTP), that due to various politically sensitive materials in the text, the China Edition has not met with the approval of the publishing authorities in China, and as such PTP will not be able to proceed with the publication of this edition. We will therefore have to cancel plans for the China Edition of your book.
He also noted that
Xiao-Li Meng told me that in China they didn’t teach Bayesian statistics because the idea of a prior distribution was contrary to Communism (since the “prior” represented the overthrown traditions, I suppose). Or maybe he was pulling my leg, I dunno.
This actually sounds more like the China of 1965 than of 2011.
A commenter proposed an alternative explanation:
It might just be that the book’s chinese publisher has been targeted by the political elite or failed to bribe someone important, or that the book is in competition with a well-connected local author’s opus, etc. Plenty of good old corruption and graft reasons that have nothing to do with the book per se.
I took a year-long course in theoretical statistics back at Stanford in 1975-76. We covered both Bayesian and frequentist theories. I still have my notes and my textbook. I never considered the political implications and have great difficulty taking this seriously.
Via BoingBoing, with Cory Doctorow’s comment:
I have no idea if it’s true that China prohibits Bayesian math, but if they do, I wonder what they use for spam-filtering (not to mention censorware).