PEMODELAN REGRESI HURDLE POISSON DALAM MENGATASI EXCESS ZEROS UNTUK KASUS PENYAKIT TETANUS NEONATORUM PADA NEONATAL DI JAWA TIMUR

*Cylvia Evasari Margaretha  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Dwi Ispriyanti  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Tatik Widiharih  -  Departemen Statistika, Fakultas Sains dan Matematika, Universitas Diponegoro, Indonesia
Published: 30 Aug 2019.
Open Access Copyright 2020 Jurnal Gaussian
License URL: http://creativecommons.org/licenses/by-nc-sa/4.0

Citation Format:
Abstract

Tetanus Neonatorum is one of the infectious diseases that occur in newborns caused by Clostridium Tetani bacteria through cuts or scratches. The number of Tetanus Neonatorum cases in East Java Province in 2017 is discrete data Poisson distribution with a proportion of zero value of 73,7 percent. The amount of zero value data can result in overdispersion where the variance is greater than the mean. To overcome this problem, Hurdle Poisson regression model is a solution. To estimation of regression parameters for Hurdle Poisson regression is using the Maximum Likelihood Estimation (MLE) method and Broyden Fletcher Goldfarb Shanno (BFGS) iteration. Hurdle Poisson regression produces predictor variables that affect the number of Tetanus Neonatorum cases in East Java Province in the logit model are the percentage of pregnant women administered the K4 program, population density per  and in the truncated Poisson model are the percentage of labor assisted by health workers the percentage of pregnant women administered the K4 program, population density per  with the Akaike Information Criterion (AIC) value of 78,422.

Keywords: Tetanus Neonatorum, Excess Zeros, Overdispersion, Hurdle Poisson Regression

Keywords: Tetanus Neonatorum, Excess Zeros, Overdispersion, Hurdle Poisson Regression

Article Metrics: