Forecasting the Number of Hat Yai International Airport Passengers

Abstract

The objective of this research is to compare three forecasting techniques of the Holt-Winters method with different initial values, the Bagging Holt-Winters method, and Box-Jenkins method based on the number of monthly Hat Yai International airport passengers from January 2003 to December 2019 which have both non-linear trend and seasonal variation. The data are collected by the Airport of Thailand Public Company Limited. The data are divided into 2 sets. The first set from January 2003 to December 2018 is used to construct the models and employed minimum Root Mean Square Error (RMSE) and residuals have normal distribution for model selection. The second set is from January 2019 to December 2019 which is used to compute the accuracy of forecasting models by using the Mean Absolute Percentage Error (MAPE). The results show that the additive Bagging Holt-Winters model gives the minimum RMSE = 9,0031.66 for the first set and yields MAPE = 11.66% for the second one.