GENERALIZED SPACE TIME AUTOREGRESSIVE WITH EXOGENOUS VARIABLE (GSTARX) MODELING WITH OUTLIER FACTOR ON CUMULATIVE PRECIPITATION TOWARDS INCREASING NUMBER OF DENGUE HEMORRHAGIC FEVER CASES

• Main Author: Alma Saron, Viana
• Format: Final Project
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/77879
• Institution: Institut Teknologi Bandung

Dengue Hemorrhagic Fever (DHF) is a disease caused by the Dengue virus which is transmitted through the bite of an infected female Aedes mosquito. DHF can occur throughout the year with an increase in cases usually occur in the rainy season when precipitation is high. Semarang City, which consist of sixteen administrative sub-districts, is one of the cities in Indonesia with a relatively high number of DHF cases. The increase in the number of dengue cases in sixteen sub-districts in Semarang City can be modeled using spatial time series modeling such as GSTAR and GSTARX. GSTAR models the value of observations at a location affected by the value of observations at the previous time at that location and surrounding locations. GSTARX is the expansion of the GSTAR model that considers the influence of exogenous variables that may affect the value of observations. In modeling the increase of dengue cases, cumulative precipitation is considered as an exogenous variable. This paper introduce ordinary time series modeling, GSTAR, GSTARX, and iterative procedures in detecting outliers in terms of Additive Outlier (AO) and Innovational Outlier (IO). Obtained that the GSTARX(1;1) model with Additive Outlier is a more accurate model to describe the increase number of DHF cases in Semarang City. The modeling results can be used to model the expectation of aggregate claims to BPJS Kesehatan caused by DHF cases in Semarang City.