Supriatna Adhisuwignjo, Azam Muzakhim Imammuddin, Suci Astutik
The aim of research is to determine the spatial regression model which appropriate for Dengue Hemorrhagic Fever (DHF) cases in Malang, and surrounding areas as well as creating a thematic map of the vulnerability level to Dengue Hemorrhagic Fever (DHF) using Geographic Information System (GIS). The observed variable is the effect of Larva-Free Index, rainfall, altitude, air temperature and behavior toward DHF incidence rate (IR). Regression model used is OLS Regression, Spatial Autoregressive (SAR) and the Spatial Error Model (SEM). To determine the appropriate model AIC (Akaike Information Criterion) and R-square (R2) are used. The results obtained show that the smallest AIC values obtained in the spatial regression model SEM compared to other models. SEM model provides R 2 of 0.89, which means that 89% of DHF data variability can be explained by the SEM model. Variables that significantly influence the IR DHF are people's behavior, rainfall and air temperature. Further results of the SEM model predictions described in a thematic map using GIS. Based on DHF thematic maps obtained, it can be seen that the highest level of vulnerability of the Incidence Rate (IR) was in Klojen and Lowokwaru (Malang City) and Pakis (Malang Regency). © 2012 EuroJournals Publishing, Inc.
State Polytechnic of Malang, Malang, Indonesia; Brawijaya University, Malang, Indonesia
