An estimate for the probability of heads on a coin whose relative error is independent of the true value

Start: 02/29/2016 - 4:15pm
End  : 02/29/2016 - 5:15pm

Applied Math Seminar

Mark Huber (Claremont McKenna College)


Consider a coin with an unknown probability $ p $ of heads that can be flipped as many times as needed. In this talk I will present a new estimate $ \hat p $ for $ p $ such that the relative error $ (\hat p/p) - 1 $ has a distribution that is independent of $ p $. This has applications in several Monte Carlo algorithms, including using acceptance/rejection to give approximations of high dimensional integrals and sums. In addition, this idea can also be used to obtain an estimate with similar properties for the mean of a sequence of independent, identically distributed Poisson random variables.

Millikan 1021 Pomona College