Arima Modeling of International Visitor Arrivals in the Philippines: 2008 – 2023

Abstract This descriptive, quantitative, correlational, and predictive research aimed to analyze the monthly international visitor arrivals in the Philippines from January 2008 to December 2023 and forecast future values utilizing secondary data from the Department of Tourism, encompassing 192 months, with 24 months used for forecasting. The analysis employed various statistical methods such as Time Series Analysis, Exploratory Data Analysis utilizing the Winsorization process, Autocorrelation and Partial Autocorrelation Function, Augmented Dickey-Fuller Test, corrected Akaike Information Criterion, and Bayesian Information Criterion to assess the time series components. The study identified an increasing trend in international visitor arrivals until January 2020, followed by a sudden drop in February 2020 due to the pandemic. The highest total number of international visitor arrivals in the Philippines on a monthly basis is January with 6,707,811 visitors and September obtained the lowest total number of international visitor arrivals in the Philippines with an increasing trend of 4,791,961 visitors from 2008 to 2023, irregularity was observed in the data due to pandemic. The best-fit model generated for forecasting the visitor arrivals was determined to be ARIMA(0, 1, 1)(0, 1, 3)12 exhibiting an excellent predictive accuracy based on Mean Absolute Error (MAE) of over 17,000 and Mean Absolute Percentage Error (MAPE) of 5.35% serving as metrics of forecast accuracy. The forecasted arrivals appear to be higher in 2025 compared to 2024 for each corresponding month, specifically, in 2025 with 579,805 visitors compared with 516,739 visitors on average, which may indicate an expected growth in tourism over the year. Moreover, the findings concluded that the model projects that January 2025 had the highest total number of monthly international visitor arrivals, while September 2024 had the lowest in the two-year forecast. The researchers of the study recommended the provision of updated visitor arrival data for future studies, regional data to be used in analysis, alternative forecasting techniques, and an investigation into factors affecting visitor arrivals.