← Maven Securities Interview Insights
I blanked for a second on how to even start this.
Model the process as a Markov chain with states representing the last flip (H or T) and solve for the probability that the stopping time is even. Alternatively, use symmetry and recursive equations to derive the probability directly.
Pro tip: Clearly define the states and set up equations for the probability of stopping on an even flip from each state; solving these equations is straightforward and demonstrates strong probabilistic reasoning.
Define states based on the last flip: state H (last flip was heads) and state T (last flip was tails). Let p_H be the probability that the total number of flips is even given we are in state H, and p_T for state T.
From state H, the next flip is H with probability 1/2 (stopping with one more flip, so parity flips) or T with probability 1/2 (continuing to state T). Write similar equation for state T.
Solve the two linear equations for p_H and p_T. Use symmetry to note p_H = p_T, simplifying the solution.
The process starts with the first flip, which is equally likely to be H or T. The total flips N is even if the first flip is followed by an odd number of additional flips? Actually, careful: N is the total flips including the first. If the first flip is H, then we are in state H after 1 flip, and we need the remaining flips to be odd? Let's derive properly.
Combine the initial probabilities with the solved p_H and p_T to get the overall probability that N is even.
AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.