Application of the Poisson and Negative Binomial Models to Thunderstorm and Hail Days Probabilities in Nevada

Abstract

Rare events such as thunderstorm or hail days often fit one of two distributions, the Poisson or the negative binomial. These two models were tested on monthly and annual thunderstorm days as well as on annual hail days at five locations in Nevada. A procedure for calculating efficient estimates of the parameters for the negative binomial distribution, utilizing the iterative process and the second-order polynomial model, is described. Results of the program applied to five sites in Nevada are discussed.