Identifying the Biased Coin From Limited Flips

Reported in interviews at Squarepoint Capital

Problem statement

You have two coins, one fair and one biased with $P(H) = p > \frac{1}{2}$. You don't know which is which.

You test coin $A$ by flipping it 3 times and observe 2 heads. You test coin $B$ by flipping it once and observe 1 head.

Which coin is more likely to be the biased one? Does your answer depend on the value of $p$?

Show hint

With two coins and exactly one biased, what are the two full scenarios you are choosing between, and how do Bayes' rule and the priors compare them?

Relevant roles: Data Scientist, Applied Scientist and ML Engineer.

Related free problems