Bayesian parameters estimation of the negative binomial - two parameter weighted exponential distribution with application in biological data

Abstract

The purpose of this paper is to report the results of fitting three biologically significant statistical models, Poisson, negative binomial (NB) and negative binomial-two parameter weighted exponential (NB-TWE) distributions, to biological data. The results of fitting the different distributions are compared by using the K-S Test and Cramer-von Mises Test. We found that the data follows the highest value base on a NB-TWE distribution, which implies that the NB-TWE distribution is the best fit for these data sets. Therefore, the NB-TWE distribution is the best choice for biological data that has an overdispersion problem, and are heavy-tailed with positive skewness.