When the facts change, I change my opinion. What do you do, sir? — Thomas Bayes, British mathematician and Presbyterian minister
The New York Times reviews Sharon Bertsch McGrayne’s The Theory That Would Not Die: How Bayes’ Rule Cracked the Enigma Code, Hunted Down Russian Submarines, and Emerged Triumphant from Two Centuries of Controversy.
Three topics I love to think about rolled into one: anything at all to do with Enigma, geophysical parameter estimation and the craziness behind not changing your mind given the increasing likelihood of evidence to the contrary.
336 pages long, so I kinda expect it to be a quick Winchester-esque romp through probability estimation, but any book that shows how much we use Bayes’s theorem in almost all fields of science and engineering and everyday is alright by me. In fact, Bayes is one of the first things taught in an oil and gas reservoir characterization class. Quantifying unknowns is tricky business and the subsurface is inherently unknown at best, so it is to every reservoir geophysicist’s advantage to use as many data sets as possible in parameter estimation and assign uncertainties to each input – seismic attribute volume, velocity model, core sample, log curve, etc. – as early and often as possible. (Paper: Bayesian reservoir characterization by Luiz Lucchesi Loures)
The reviewer states that “a serious problem arises, however, when you apply Bayes“s theorem to real life.” What exactly that is supposed to mean? As pointed out earlier, Bayes’s theorem is used in very real-life areas as nebulous as cryptography and the search for fossil fuels. Also, news flash: every undertaking has associated human agendas. So, why can Bayes not be implemented in studies of global climate change and autism? But on one thing we agree – the sad fact that there are many of us, scientists or not, who are “wedded to [our] priors.” So, and I guess this goes for everyone, absorb and digest as much information as possible, stop to think about or research the likelihood of what you learned and try not to let confirmation bias get in your way.
Good luck. (Get it? Good luck? Never mind.)