FORECASTING URBAN POPULATION GROWTH IN THE PHILIPPINES USING AUTOREGRESSIVE INTEGRATED MOVING AVERAGE (ARIMA) MODEL

The Philippines is one of the fastest urbanizing countries in the East Asia and Pacific region (Baker & Watanabe, 2017). Despite having its advantages, urbanization still has its challenges that require extensive urban management and development programs for it to be prevented and minimized. In this paper, the researchers forecasted the urban population growth of the Philippines using the Autoregressive Integrated Moving Average (ARIMA) Model. The historical data obtained from the World Bank Group was from 1960 to 2020. The R Programming Language was used as the medium for the entire forecasting process. Autocorrelation Function (ACF) and Partial Autocorrelation Function (PACF) plots, Augmented Dickey-Fuller (ADF) test, Phillips-Perron (PP) test, and Kwiatkowski, Phillips, Schmidt, and Shin (KPSS) test were used for testing the stationarity of the time-series data. Moreover, Akaike Information Criteria (AIC), Corrected Akaike Information Criterion (AICc), and Schwarz Information Criteria (SIC) were used as criteria for selecting the best ARIMA model. It was shown that the best ARIMA model for forecasting the urban population growth of the country is ARIMA (20, 1, 10). This model has been formulated and chosen through the mentioned statistical tests, and criteria for validation, and was further validated using error measures. The chosen ARIMA model was proven to be accurate based on the Root Mean Square Error (RMSE) of 0.18877 and the Mean Absolute Percentage Error (MAPE) of 3.71%. The researchers found an increase in the trend of 1.95% by 2022, 2.08% by 2024, 2.19% by 2026, and 2.36% by 2028. This potential rise in urban population growth in the Philippines may improve the economy of the country for the next 6 years, but this could also imply that the underlying issues of urbanization may get worse. The researchers conclude that the Philippine national government and local government units should have better and strengthened urban management and development programs to aid these problems. Government officials and even private sectors may use this paper as a reference to have an informed decision and policy-making. KEYWORDS: Autoregressive Integrated Moving Average (ARIMA) Model, Box-Jenkins Method, Urbanization, Urban Population Growth, Forecast, R Programming Language