Hybrid Ensemble Spatial Regression Model for Case Number of Dengue Hemorrhagic Fever in Central Java Province

摘要

Spatial regression model is a regression model that is formed because there is relationship among observations on the dependent variable or among its errors. One of assumptions in spatial regression model is homogeneous of error variance, but we often find the diversity of data in several different locations so that the assumption is not met. The objective of this research is to get the best model if we find the heterogeneity in error. For the application, we use data of dengue hemorrhagic fever (DHF) patients in Central Java Province. Ensemble technique is done by simulating noises from a normal distribution with mean zero and standard deviation $sigma $ of the error spatial model and adding noise to the dependent variable. The result on the case of a number of DHF patients in Central Java Province showed that there were effects of lag and error spatial dependence so that we use spatial autoregressive model and spatial error model. Because these two spatial models are met, we then use a hybrid ensemble spatial regression model. As a result, the hybrid ensemble spatial regression model is the best model because it has the smallest root mean square error (RMSE) and has no heterogeneity in variance.