Bayes Bogus?

Mr. Spock, of the original Star Trek, was pretty much the perfect straight man. He never got the joke, and he never got the girl. Which is why, I suppose, that Jim kept him on despite his limitations as a first officer. I am thinking here of his complete incompetence at elementary probability calculations. Whenever some dangerous mission was contemplated, Spock would portentiously announce some six or eight digit probability of failure, with success being somewhere out there in tens or hundred thousanths of a percent. Undeterred, Kirk would set out, usually taking with him the pilot, the ship's doctor, and the only key to the officers' bathroom. Naturally Spock was always wrong, but nobody ever complained or sent him back to remedial statistics.

I'm sure he must have somehow become confused about Bayesian inference. Using Bayes theorem to compute probabilities is of course perfectly legitemate and absolutely necessary for most statistical reasoning, but the Bayesian interpretation of probability drives me nuts. From the Wikipedia:
There are two broad categories of probability interpretations: Frequentists talk about probabilities only when dealing with well defined random experiments. The relative frequency of occurrence of an experiment's outcome, when repeating the experiment, is a measure of the probability of that random event. Bayesians, on the other hand, assign probabilities to any statement whatsoever, even when no random process is involved, as a way to represent its subjective plausibility.
Subjective plausibility makes my skin crawl. I have no idea what the hell it is supposed to mean. I can't think of any case when an expressed probability isn't either based on an explicit or implicit frequentist interpretation or else, like Mr. Spock's invariably misguided predictions, merely an ignorant prejudice.

Consider James Annan's
discussion of Bayesian probabilities referenced earlier. He talks about the "probability" that the number "1,234,567,897" is prime. Of course at this point I know it's zero, since it's divisible by 994817 as well as two two digit primes, but he discussed some general methods for guessing the probability of primality for an arbitrary 10 digit integer based on things like the asymptotic density of primes (number of primes less than x is asymptotic to x/(ln x)), as well as the simple tests for divisibility by 2, 3, and 5. Each of these estimates is based on a frequency interpretation. I don't see any room for any interpretation that isn't essentially frequentist.

The other essential element is theory. Without some theory connecting previously observed events with future events, no prediction is possible.

Comments

Post a Comment

Popular posts from this blog

Anti-Libertarian: re-post

The most pointed critique of social security and liberalism in general is the libertarian critique. The problem with libertarians, for a liberal, is that we have too much in common. We both believe in individual rights, tolerance of individual differences, and dislike government prescription of religion. The basic difference, it seems to me, is the different answers we give to Cain's (with thanks to the Captain - my original version had the corpse asking the question!) famous question: "Am I my brother's keeper?" That's not really my basic bitch against the libertarians though. My real complaint is the same as my complaint against most religion - its premise is a fraud. For those who can't stand to wait for the punchline, I believe that trying to implement libertarian principles leads to tyranny or social disintegration. Demonstrating that takes some historical (and pre-historical) context. For all but the last 15,000 or so of the 100,000 years the human spec
Read more

Uneasy Lies The Head

Uneasy lies the head that wears a crown ........... Shakespeare Henry IV, Part II. So it is with us and our slightly distant primate cousins , so it seems. A 9-year study of wild baboons in Kenya by Princeton University and the Institute of Primate Research may turn conventional assumptions of alphas males upside down. The study found that alpha males have high testosterone levels, which allows them to dominate other baboons, get access to more food, and attract better mates. This is expected. Surprisingly, though, the alpha males also have high levels of stress hormone glucocorticoid, which scientists measured from their feces samples. In fact, the stress of alpha baboons was on par with low-ranking baboons. The convention wisdom is that access to food and females provides life security and thus less stress for dominant males. Conversely, weak males who are constantly threatened by starvation and physical violence from stronger males are thought to be more stressed. The st
Read more

We Call it Soccer

The US played its best, and maybe only good game of World Cup, but lost thanks to brilliant goalkeeping by the Swedish keeper and the bizarre method soccer uses to settle ties when three US stars could not put the ball on frame from the 12 yard spot.  At the end it came down to a blocked shot that spun in by maybe a couple of centimeters. Well, that's football - too much like life.   Which brings me to a modest proposal.  Replace the thirty minute overtimes with as many 7-player 15 minute periods as needed to settle it.  Scoring should be much easier with fewer players.
Read more