A forecast of Singapore’s tourism arrivals by air using time series analysis

Abstract

Tourists from Indonesia, China and Malaysia rank top three among Singapore’s overall inbound tourists. This research develops and compares three univariate time series models to forecast short-term tourist arrivals from these three countries. Based on the historical data over the period from January 1995 to December 2011 for Malaysia, and from January 1998 to December 2011 for Indonesia and China, Autoregressive Integrated Moving Average (ARIMA), Seasonal ARIMA (SARIMA) and ARIMA-Generalized Autoregressive Conditional Heteroskedasticity (ARIMA-GARCH) models are employed. The out-of-sample forecasting performance of these best-fitting models are investigated further over the period from January 2012 to June 2013, according to the root mean squared error (RMSE), mean absolute percentage error (MAPE) and Diebold-Mariano (DM) Statistic. The optimal model is applied to obtain post-sample forecasts for the next six months. ARIMA proves to be the optimal model for all three countries. This paper will not only provide valuable information to aviation and tourism sectors, but also boost interest to forecast international travel demand for Singapore.