2023/09/23 by Tanner C. Reese, Reese, Tanner
Computer Science · Decision Sciences · Mathematics · #06A05 #06A06 #60-XX #Bayesian Methods and Mixture Models #Combinatorics (math.CO) #FOS: Mathematics #Probability (math.PR) #Probability and Risk Models #Stochastic processes and statistical mechanics
paper · pdf · doi:10.48550/arxiv.2309.13250
openalex publication_date 2023/09/23 · openalex created_date 2023/09/27 · openalex updated_date 2026/07/28
We determine the distributions of lengths of runs in random sequences of elements from a totally ordered set (total order) or partially ordered set (partial order). In particular, we produce novel formulae for the expected value, variance, and probability generating function (PGF) of such lengths in the case of an arbitrary total order. Our focus is on the case of distributions with both atoms and diffuse (absolutely or singularly continuous) mass which has not been addressed in this generality before. We also provide a method of calculating the PGF of run lengths for countably series-parallel partial orders. Additionally, we prove a strong law of large numbers for the distribution of run lengths in a particular realization of an infinite sequence.