Forecasting airport passenger traffic: the case of Hong Kong International Airport

Abstract

Hong Kong International Airport is one of the main gateways to Mainland China and the major aviation hub in Asia. An accurate airport traffic demand forecast allows for short and long-term planning and decision making regarding airport facilities and flight networks. This paper employs the Box-Jenkins Autoregressive Integrated Moving Average (ARIMA) methodology to build and estimate the univariate seasonal ARIMA model and the ARIMX model with explanatory variables for forecasting airport passenger traffic for Hong Kong, and projecting its future growth trend from 2011to 2015. Both fitted models are found to have the lower Mean Absolute Percentage Error (MAPE) figures, and then the models are used to obtain ex-post forecasts with accurate forecasting results. More importantly, both ARIMA models predict a growth in future airport passenger traffic at Hong Kong.