Some years ago, the late Roger Farthing, a neighbour and the author of a definitive history of Mount Sion, the area where I live in Tunbridge Wells, rang me up to ask if I had heard of Thomas Bayes. He was researching his book at the time and had discovered that Bayes once lived near us. I had not heard of him, and until the 1950s, neither had many other people. This plaque was introduced to coincide with the fourth centenary of Tunbridge Wells only a couple of years ago. It was not until the era of computers that Bayes' theory of probability came to be appreciated outside the narrow world of specialist mathematics. Since then, though not every statitician is a Bayesian, the theory has proved useful in many areas of forecasting and statistical analysis. Essentially, Bayes introduced into the analysis of probablities, existing and future beliefs and degrees of uncertainty, as distinct from assigning probabilties to random events according to their frequency of occurrence. The theorem, which explains the theory is beyond me, but it appeals because it brings into play, in the cold field of statistics, a strongly human, slightly messy element. Oh yes, and close to some of our own preoccupations, Bayes' theory is now applied in the tricky business of filtering email spam.
Around this date, the lady in the building society usually asks me if I have done my Christmas shopping. Today, as I approach the counter and see her friendly face coming towards me, I find myself wondering whether she will be true to form. At first there is nothing, then as she hands my pass bookbook back, she asks, "are your ready for Christmas?" A pleasing variation, but then, it is a bit early to have have done any Christmas shopping. Perhaps if I go to the building society again before Christmas... Who knows? I have reached the age when changes are seldom welcome.
In the window of the new butcher's shop in Chapel Place, is a dish full of cooked bones. "What are those for? " asks Heidi. "Postman's ankles, " says the butcher, " they're for dogs," he explains and adds: "It's our attempt at a sense of humour."
You'll remember my recent enthusiastic post on "The Drunkard's Walk" a book about randomness by Leonard Mlodinow. Here's a quote:
"(Thomas Bayes invented) a theory of 'conditional probability... For example the probability that a randomly chosen person is mentally ill and the probability that a randomly chosen person believes his spouse can read his mind are both very low. But the probability that a person is mentally ill if he believes his spouse can read his mind is much higher, as is the probability that a person believes his spouse can read his mind if he is mentally ill."
As a newcomer to Thomas Bayes, all of this sounds incredibly modern. Sometimes people on holiday unexpectedly bump into people they know from home. The modern science of sociology sees this as a highly predictable occurrence not a strange co-incidence. It would seem that Bayes predicted the sociologists as well.
I like the idea of factoring in future beliefs. Sounds almost Alice in Wonderlandish - as does Postman's ankles. A delightful post.
Thanks for comments. I had wondered, BB, whether Thomas Bayes featured in The Drunkard's Walk. The quote very well illustrates the appeal of the theory.
Post a Comment