Approximation of the Median of Poisson Distribution Using the Relation of Poisson-Gamma Distribution

Abstract

A major drawback of median is that median cannot usually be defined in closed form, even in the cases where their distribution functions are known. This study gives an approximation of the closed form for the median of the Poisson distribution by using the relation of Poisson-Gamma distribution. The approximation uses elementary techniques based on the monotonicity of certain sequences involving tail probabilities of the Poisson distribution and the Central Limit Theorem. Furthermore, a closed form expression for the mean absolute deviation 􀜧|􀜰􀰒 􀵆 􀜯􀝁􀝀􁈺􀜰􀰒􁈻| is also studied. The median is relevant because of the fact that it minimizes the mean absolute deviation of a Poisson random variable (􀜰􀰒􁈻 by considering median as central tendency.