Abstract:
The objective of this research is to study and compare two models of forecasting time series, namely SARIMA and RegSARIMA models. In this study, we use the monthly time series data of number of air passengers of Suvarnabhumi airport and time series of independent variables, namely time, seasonal index, irregular variation index and air passenger numbers at time t-12. The data used was monthly time series from January 2003 to December 2017. The Mean Absolute Percentage Error (MAPE) and Adjusted R-Squared are used as the comparative criteria. It was found that the RegSARIMA((1,4),0,(1,23))(0,0,1) is the most suitable model for forecasting the monthly time series data of number of air passengers of Suvarnabhumi airport. The RegSARIMA model had the lowest MAPE of 2.7034 percent with the highest Adjusted R-Squared of 98.1100 percent . Predicted air passenger numbers of Suvarnabhumi airport in 2018, 2019 and 2020 are 62,032,824, 64,599,553 and 67,871,705 persons, respectively.The number of air passenger at Suvarnabhumi airport is predicted to increase more than 67 million persons within three years.