ESTIMASI PARAMETER DISTRIBUSI BINOMIAL NEGATIF MENGGUNAKAN METODE INFERENSI BAYESIAN

Laura Erlinda Laidat
Keristina Br Ginting
Ganesha Lapenangga Putra

Jurnal Diferensial
Vol 4 No 1 (2022): April 2022

Parameter estimation is a form of inferential statistics. Parameter estimation consists of point parameter estimation and interval parameter estimation. In this study, point and interval parameter estimation of the Negative Binomial distribution will be carried out using the Bayes method. The Bayes method in this study utilizes the Beta distribution as the conjugate prior, the Uniform distribution as the non-conjugate prior, and the Jeffrey method as the non-informative prior. To evaluate the best estimator, the method used is to look at the smallest value of the posterior variance and the width of the credible Bayes interval. In a simulation study using R programming, the best estimator is the beta conjugate prior, because it has the smallest posterior variance value and the smallest credible Bayes interval width compared to the Uniform non-conjugate prior and Jeffrey's non-informative prior.