Bayes and the Fiddler

An ex se intellegitur from The Fiddler

You are given an urn containing 100 balls. N of them are red, and 100−N green, where N is chosen uniformly at random between 0 and 100 (inclusive). You take a random ball out of the urn—it just so happens to be red—and discard it. Is the next ball you pick, from among the 99 remaining balls, more likely to be red or green?

since this is Bayes or Laplace in action. Namely, since P(R,R⁰|N)=N(N-1)/100×99 and

P(R^1|R^0)=\sum_{N=0}^{100} P(R^1,N|R^0)

or

\sum_{N=0}^{100} P(R^1,R^0|N) \big/ \sum_{N=0}^{100} P(R^0|N)

the probability is,

1/99 \sum_{N=0}^{100} N(N-1) \big/ \sum_{N=0}^{100} N

which is equal to  2/3 (for any total number of balls) since

n(n+1)(2n+1)/6-n(n+1)/2=n(n+1)(2n-2)/6

by a variation of Gauss formula. Once again ChatGPT³ got it all wrong:

Leave a Reply

Discover more from Xi'an's Og

Subscribe now to keep reading and get access to the full archive.

Continue reading