Pencegahan Penyakit Kusta di Lingkungan Hutan Tropis Lembab Kalimantan Melalui Pemodelan Geographically Weighted Poisson Regression
Geographically Weighted Poisson Regression (GWPR) model is a regression model developed from Poisson regression which is applied to spatial data. Parameter estimation of the GWPR model is done at each observation location using spatial weighting. This study goal is to obtain the GWPR model and the factors influencing the number of leprosy cases in each regency(municipality) on Kalimantan Island in 2018. Spatial weighting was obtained by using the adaptive bisquare kernel function and optimal bandwidth was determined by using Generalized Cross-Validation (GCV) criteria. The data of this study was secondary data namely the number of leprosy cases in 56 regency on Kalimantan Island in 2018. The parameter estimation method of GWPR model is Maximum Likelihood Estimation (MLE). The results of analysis showed that maximum likelihood estimator is obtained by using the Newton-Raphson iterative method and the factors affecting the number of leprosy cases in each regency were different and locally. The factors influencing locally were the number of health facilities, the number of health workers, the number of male population and population density.