Monday, June 27, 2016

Brexit: "Bayesian" statistics renamed "Laplacian" statistics

With the U.K. leaving the E.U., it's time for "Bayesian" to exit its titular role and be replaced by "Laplacian".  

;-)

Various historians (e.g., Dale, 1999; McGrayne, 2011; as cited in DBDA2E) have argued that despite Bayes and Price having published the basic formula first, it was Laplace who did all the heavy lifting to develop the methods. Laplace was French, while Bayes and Price were English. 

On the other hand, Bayes did attend the University of Edinburgh, and lived around London, so maybe he would have voted to stay!

This humorous implication of Brexit was suggested by a participant at a workshop I presented in St. Gallen, Switzerland, last week, after I briefly described the history above and showed the book photographed at Bayes' tomb and at Laplace's memorial. I have forgotten who it was that made the joke. If the originator of the joke is reading this, please put a comment below or send me an email!

4 comments:

  1. Ah but Bayes and Price, I assume, achieved at least a relatively high educational level (pardon my ignorance) and so http://www.theatlantic.com/international/archive/2016/06/brexit-vote-statistics-united-kingdom-european-union/488780/ could have been predicted to vote remain.

    Your article may be rather tongue-in-cheek but it would be interesting to take this analysis futher. With so much analysis of the effect of demographics on votes, and what voters now feel about the result, perhaps we could predict the outcome of a second referendum?

    Yours, clinging on to hope, a remain voter...

    ReplyDelete
  2. @Isabel Sargent:

    Thanks for your comment. I agree. :-)

    (Along those lines, that's why I mentioned, in the post: "Bayes did attend the University of Edinburgh, and lived around London, so maybe he would have voted to stay!" Brexit voters in Scotland and around London strongly favored to remain.)

    ReplyDelete
  3. Probably it is time to accept that Calculus is invented by Leibniz.

    ReplyDelete
  4. I was the one who made the joke.
    Maybe my suggestion is slightly biased because I have a French passport. But I am also German and as a German I am pro Laplace, too.
    On the other side, the genealogical table of great mathematicians is full of French names. Even without Laplacian statistics the French contribution to the discipline is undisputed.
    And the Brexit is probably not the end of the story. Let´s go on with "Bayesian" until they come back, in one way or another.

    ReplyDelete