Application of Geographically Weighted Bivariate Poisson Inverse Gaussian Regression

Abstract

West Sumatera is a region with low leprosy cases. On the other hand, it is worrying because it has a significant increase. The number of Pauci Bacillary (PB) and Multi Bacillary (MB) leprosy in West Sumatera is one of count data within over dispersion so that can be modeled by Geographically Weighted Bivariate Poisson Inverse Gaussian Regression (GWBPIGR). This study will be developing the regression model that has geographic weighting factors, which will provide different local models in each location. The local models are influenced by several factors such as geographic conditions, social, culture and economic. In GWBPIGR model, parameter estimation obtained by Maximum Likelihood Estimation (MLE) method with Newton-Raphson algorithm. Meanwhile, hypothesis testing is acquired by Maximum Likelihood Ratio Test (MLRT) method. Explanatory variables that influence new cases of PB and MB leprosy in district of West Sumatera 2014 are as following; percentage of healthy house, percentage of poor people, percentage of hygiene food processing place, and ratio of medical personnel.