Model Geographically Weighted Weibull Regression pada Indikator Pencemaran Air Biochemical Oxygen Demand di Daerah Aliran Sungai Mahakam
Abstract
Geographically Weighted Weibull Regression (GWWR) Model is a Weibull regression model applied to spatial data. Estimation of the GWWR model is performed at every observation location using spatial weighting. The purpose of this study was to determine the GWWR model of water pollution indicator Biochemical Oxygen Demand (BOD) data and the factors that influence BOD in the Mahakam River. The estimating parameters method of the GWWR model was the Maximum Likelihood Estimation (MLE) and it’s estimator was obtained by Newton-Raphson Iterative method. Spatial weighting in parameter estimation was determined using the Adaptive Bisquare weighting function and bandwidth optimum was determined by using Generalized Cross-Validation (GCV) criteria. Based on the GWWR model parameters testing, the factors that influence BOD locally was nitrate concentrations, while the factors influence globally were temperature and nitrate concentration.