A note on maxima in random walks (Q907270)

Summary: We give a combinatorial proof that a random walk attains a unique maximum with probability at least \(1/2\). For closed random walks with uniform step size, we recover Dwass's count of the number of length \(\ell\) walks attaining the maximum exactly \(k\) times. We also show that the probability that there is both a unique maximum and a unique minimum is asymptotically equal to \(\frac{1}{4}\) and that the probability that a Dyck word has a unique minimum is asymptotically \(\frac{1}{2}\).