POISSON REGRESSION OF DAMAGE PRODUCT SALES USING MCMC

Reny Rian Marliana (STMIK Sumedang)
Septiadi Padmadisastra (Department of Statistics, Padjadjaran University, Indonesia)

Indonesian Journal of Statistics and Its Applications
Vol 2 No 1 (2018)

In this paper a model for the number of “damage” product sales is studied. The product sales are run into underreporting counts, caused by a delay on input process of the system called sales cycle. The goal of the study is to estimate the parameters of the regression model of product sales on an explanatory variable. It is the actual number of product sales. The model used is a mixture of the Poisson and the Binomial distributions. The parameters of the regression model are estimated by a Bayesian approach and Markov Chain Monte Carlo simulation using Gibbs sampling algorithm. The results of estimation clearly showed a gap between undamage product sales and the actual number. The gap is the number of damaged product sales.