rise of the B word
While preparing a book chapter, I checked on Google Ngram viewer the comparative uses of the words Bayesian (blue), maximum likelihood (red) and frequentist (yellow), producing the above (screen-copy quality, I am afraid!). It shows an increase of the use of the B word from the early 80’s and not the sudden rise in the 90’s I was expecting. The inclusion of “frequentist” is definitely in the joking mode, as this is not a qualification used by frequentists to describe their methods. In other words (!), “frequentist” does not occur very often in frequentist papers (and not as often as in Bayesian papers!)…
February 26, 2013 at 7:40 pm
More interesting is this set of graphs:
http://books.google.com/ngrams/graph?content=bayesian%2C+prior+distribution%2Cposterior+distribution%2Cp-value%2Ct-test&year_start=1920&year_end=2008&corpus=15&smoothing=3&share=
It seems that people have started publishing more about statistics, but the relative rate of growth of the different terms seems indicative that in publishing, bayesian methods are still WAY behind, in terms of times they are mentioned.
February 27, 2013 at 12:05 pm
Thank you, David: while I can spot the difference in numbers, I am not so sure the comparison of the absolute numbers makes so much sense: when one uses “p-value” in a paper or in a book, it is likely to be used several times (a few thousand times in the extreme cast of The Cult!). I would think the use of “Bayesian” would remain more limited. Anyway, I do not want to start an endless debate: my aim was to try to spot the impact of the introduction of MCMC methods in 1990 on the use of the term “Bayesian” and it does not appear on those graphs.
March 2, 2013 at 5:49 am
Google’s indexing is case-sensitive, and you get very different results if you try with a capital B: http://books.google.com/ngrams/graph?content=Bayesian%2Cprior+distribution%2Cposterior+distribution%2Cp-value%2Ct-test&year_start=1920&year_end=2008&corpus=15&smoothing=3&share=
March 2, 2013 at 7:23 am
Oh, truly?! That indeed makes a whole difference.
February 26, 2013 at 2:03 pm
Google Trends is much more pessimistic than Google N Grams :
http://www.google.com/trends/explore#q=bayesian,%20frequentist,%20maximum%20likelihood
Google Trends index is a measure of search intensity on google, not of the use of the word in books and/or articles.
By the way, it is not clear if recent scientific publications are included in Google N Grams.