Title: Human Immunodeficiency Virus(HIV) Cases in the Philippines: Analysis and Forecasting

Abstract: Reports from the Health Department in the Philippines show that cases of Human Immunodeficiency Virus (HIV) are increasing despite management and control efforts by the government. Worldwide, the Philippines has one of the fastest growing number of HIV cases. The aim of the study is to analyze HIV cases by determining the best model in forecasting its future number of cases. The data set was retrieved from National HIV/AIDS and STI Surveillance and Strategic Information Unit (NHSSS) of the Department of Health containing 132 observations. This data set was divided into two parts, one for model building and another for forecast evaluation. The original series has an increasing trend and is nonstationary with indication of non-constant variance. Box-Cox transformation and ordinary differencing were performed on the series. The differenced series is stationary and tentative models were identified through ACF and PACF plots. SARIMA has the smallest chosen AIC value. The chosen model undergoes the diagnostic checking. The residuals of the model behave like a white noise while the forecast errors behave like a Gaussian white noise. Considering all diagnostics, the model may be used for forecasting the monthly cases of HIV in the Philippines. Forecasted values show that HIV cases will maintain their current trend.