PENERAPAN REGRESI ZERO-INFLATED NEGATIVE BINOMIAL (ZINB) UNTUK PENDUGAAN KEMATIAN ANAK BALITA

Abstract

One method of regression analysis used to analyze the count data is Poisson regression. Poisson regression requires that the mean value equal to the value of variance (equidispersion). However, sometimes the data is going overdispersion  the state variance values ??greater than the mean value. One of the causes overdispersion is the excessive number of zero values ??on the response variable (excess zeros). One method of analysis that can be used on data that had overdispersion due to excess zeros is regression Zero-Inflated Negative Binomial (ZINB). The data that can be analyzed using the ZINB regression is the early childhood mortality in the province of Bali because much of the data is zero. The analysis showed that the data had overdispersion on Poisson regression, so the ZINB regression analysis was used. From the results of the ZINB regression can overcome overdispersion so it was better than the Poisson Regression Model.