DEVELOPMENT OF THE UNIVARIATE TIME SERIES MODEL FOR FORECASTING DENGUE HEMORRHAGIC FEVER CASES IN NAKHON SI THAMMARAT

Abstract

Dengue Hemorrhagic Fever (DHF) is a prominent cause of hospitalization and death in Thailand, especially in the south. Epidemiological modeling was used to estimate the trend of outbreak tendencies based on epidemic data. The goal of this work was to create a realistic model for estimating DHF occurrences using monthly data from the department of disease control, ministry of public health, Thailand from 2010 to 2019. SARIMA model with the Box-Jenkins approach was conducted to forecast dengue incidence using the previous data. Bayesian Information Criteria (BIC), Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) were used to determine their accuracy. The result shows that the SARIMA(2,0,1)(1,0,0)12 model fits the Nakhon Si Thammarat pandemic data the best. Their accuracy had the smallest BIC, MAPE, and RMSE yielding 9.64, 848.743, and 214.661, respectively. The DHF ARIMA model is necessary and may be used to forecast the incidence of DHF in other locations as well as help in the development of public health initiatives to prevent and treat the condition.