Convex hulls of random walks: Expected number of faces and face probabilities
Abstract
Consider a sequence of partial sums Si= 1+…+i, 1≤ i≤ n, starting at S0=0, whose increments 1,…,n are random vectors in Rd, d≤ n. We are interested in the properties of the convex hull Cn:=Conv(S0,S1,…,Sn). Assuming that the tuple (1,…,n) is exchangeable and a certain general position condition holds, we prove that the expected number of k-dimensional faces of Cn is given by the formula E [fk(Cn)] = 2· k!n! Σl=0∞[]0ptn+1d-2l \\0ptd-2lk+1, for all 0≤ k ≤ d-1, where []0ptnm and \\0ptnm are Stirling numbers of the first and second kind, respectively. Further, we compute explicitly the probability that for given indices 0≤ i1<… <ik+1≤ n, the points Si1,…,Sik+1 form a k-dimensional face of Conv(S0,S1,…,Sn). This is done in two different settings: for random walks with symmetrically exchangeable increments and for random bridges with exchangeable increments. These results generalize the classical one-dimensional discrete arcsine law for the position of the maximum due to E. Sparre Andersen. All our formulae are distribution-free, that is do not depend on the distribution of the increments k's. The main ingredient in the proof is the computation of the probability that the origin is absorbed by a joint convex hull of several random walks and bridges whose increments are invariant with respect to the action of direct product of finitely many reflection groups of types An-1 and Bn. This probability, in turn, is related to the number of Weyl chambers of a product-type reflection group that are intersected by a linear subspace in general position.
Turn this paper into a lesson
ArcXiv compiles a structured reading guide from this paper's metadata: plain-English importance, contributions, prerequisite concepts, which sections to read first, flashcards, and a quiz. Grounded in the abstract, never invented.