Skip to main content

I'm safe: I'm on a mission from God











P(H/D) = [P(D/H) x P(H)] ÷ P(D).


That is one common formulation of Bayes' theorem, and at the heart of the view of statistics associated with a mid-18th century Englishman, Thomas Bayes.


I blogged about this in January. If you don't understand the notation above, and would like a primer, go here.


I bring it up again because Bayesianism has attracted new attention in the blogosphere.


Sometimes the above is called "Bayes' rule," because it suggests a rule for constantly updating your views of the likelihood of events. Your hypothesis of probability of H on Tuesday (the prior) is to be updated according to what happens or doesn't happen Wednesday, thereby yielding the posterior probability.


Econoblogger Noah Smith wrote a post he calls "Bayesian Superman." He combines a (male) teenager's sense of invincibility with something like Pascal's wager, and proposes the following thought experiment:


Suppose that you believe that there is a nonzero probability that H is true, where H in this case is the hypothesis, "God is watching out for me, has a special purpose in mind for me, and so whatever the apparent dangers, He will not allow my death prior to the accomplishment of that purpose."


What Smith is suggesting first is that H may have been adopted however you like, perhaps with a wagering element to it. After all, although today's working hypothesis is a modification of yesterday's prior, and yesterday's prior was at some point a posterior after the modification of the previous day's prior ... acknowledging all that, this cannot be an infinite string in any particular life. There was a first H, the prior of priors.


So we get to the tricky part. You are a faithful Bayesian, updating your hypotheses day by day. You did not die yesterday. You did not die the day before. And so forth. If you have entertained H as a hypothesis for many days, then presumably your confidence in H has increased with each sunset, where the relevant datum, D, is simply "I didn't die today".


So, we get somewhat further than Pascal would allow here. Not only might Pascal argue that the original wager H is rational, but Bayes would butt in to say that this ever increasing confidence in it is likewise rational.


Noah Smith wrote a think piece -- some of the above is a rather loose paraphrasing of it -- in order to explore the "observed behavior of teenagers," who seem to be Bayesians by instinct. Another blogger, Lars P. Syll, has picked up on the thought experiment and usef it as an attack on Bayesianism, which he describes as an internally consistent system of thought that nonetheless threatens harm to science.

Smith, on his twitter account, says that he didn't really intend it the way Syll took it, but he does think Syll makes a good point.


There is a lot to think about here. One obvious problem with Smith's thought experiment, though, is survivorship bias. Starting with secularist premises if I may: only those who have not yet died can be said to draw inferences at all, Bayesian or otherwise.  Someone who has died doesn't then have the luxury of saying, "aha! there is now a zero probability of H."


Smith notices the survivorship issue, then waves his hands a bit around it.


I suspect (a related point) that Bayes would have had a difficulty with premises that need the first-person pronoun "I" or "me" to make any sense. Certainly lemming C can notice that his friends, lemmings A and B, both died after jumping off yonder cliff, and that would decrease to a rational lemming the plausibility of any hypothesis that makes the act of jumping off a cliff seem a safe one. The fact that A and B can't be aware of this inference shouldn't  keep C from drawing it.


Whatever one's take on God, Ubermensch, or teenagers, I don't think that Smith's meditation makes much of a dent in the case for Bayesian statistics, under any non-parodic understanding of the latter.







Comments

Post a Comment

Popular posts from this blog

England as a Raft?

In a lecture delivered in 1880, William James asked rhetorically, "Would England ... be the drifting raft she is now in European affairs if a Frederic the Great had inherited her throne instead of a Victoria, and if Messrs Bentham, Mill, Cobden, and Bright had all been born in Prussia?"

Beneath that, in a collection of such lectures later published under James' direction, was placed the footnote, "The reader will remember when this was written."

The suggestion of the bit about Bentham, Mill, etc. is that the utilitarians as a school helped render England ineffective as a European power, a drifting raft.

The footnote was added in 1897. So either James is suggesting that the baleful influence of Bentham, Mill etc wore off in the meantime or that he had over-estimated it.

Let's unpack this a bit.  What was happening in the period before 1880 that made England seem a drifting raft in European affairs, to a friendly though foreign observer (to the older brother…

Cancer Breakthrough

Hopeful news in recent days about an old and dear desideratum: a cure for cancer. Or at least for a cancer, and a nasty one at that.

The news comes about because investors in GlaxoSmithKline are greedy for profits, and has already inspired a bit of deregulation to boot. 

The FDA has paved the road for a speedy review of a new BCMA drug for multiple myeloma, essentially cancer of the bone marrow. This means that the US govt has removed some of the hurdles that would otherwise (by decision of the same govt) face a company trying to proceed with these trials expeditiously. 

This has been done because the Phase I clinical trial results have been very promising. The report I've seen indicates that details of these results will be shared with the world on Dec. 11 at the annual meeting of the American Society of Hematology. 

The European Medicines Agency has also given priority treatment to the drug in question. 

GSK's website identifies the drug at issue as "GSK2857916," althou…

Francesco Orsi

I thought briefly that I had found a contemporary philosopher whose views on ethics and meta-ethics checked all four key boxes. An ally all down the line.

The four, as regular readers of this blog may remember, are: cognitivism, intuitionism, consequentialism, pluralism. These represent the views that, respectively: some ethical judgments constitute knowledge; one important source for this knowledge consists of quasi-sensory non-inferential primary recognitions ("intuitions"); the right is logically dependent upon the good; and there exists an irreducible plurality of good.

Francesco Orsi seemed to believe all of these propositions. Here's his website and a link to one relevant paper:

https://sites.google.com/site/francescoorsi1/

https://jhaponline.org/jhap/article/view/3

What was better: Orsi is a young man. Born in 1980. A damned child! Has no memories of the age of disco!

So I emailed him asking if I was right that he believed all of those things. His answer: three out of …