Forecasting acute respiratory infection cases in Southern Bogota: EARS vs. ARIMA and SARIMA

Acute Respiratory Infection (ARI) diseases are a major cause of hospitalization and death worldwide. They are the leading cause of morbidity in developed countries and the leading cause of death in developing countries. The incidence of such diseases constitutes a large portion of the reported cases from medical consultations in southern Bogota. This study analyzes the reported individual attention records of Hospital del Sur
¹
. The ARI-related epidemic information gathered from 2009 to 2014 have been imported by an ad hoc Extraction, Transformation and Loading (ETL) process to take them from their original raw text representation to a relational model in SQL (PostgreSQL). This study applies and compares five different models (EARS algorithms C1, C2 and C3, as well as, ARIMA and SARIMA models) to make forecast from ARI-related time series data. We show for this type of time series ARIMA model achieved the best results. Finally, we discuss about the obtained results and present conclusions and future work.