Inferência estatística para o processo geométrico baseado na distribuição Birnbaum- Saunders Reparametrizada

Abstract

In this monography, we investigate the problem of statistical inference for the geometric process (GP) when the distribution of the first occurrence time is Reparameterized Birnbaum-Saunders (RBS). We propose a model entitled GP-RBS capable of acting in this problem. Maximum likelihood estimators (EMV), estimators by the moment method (MME) and modified moment estimators (MM) are obtained. In addition, we discuss some important asymptotic properties of these estimators, such as asymptotic distribution and consistency. Next, we conduct some studies of Monte Carlo simulations with different parameter values to compare the performance of the results obtained and, finally, we present the applications in four real data sets.