초록 열기/닫기 버튼
The purpose of this study is to select the most appropriate quantitative forecasting model to predict the Japanese travel demand to Korea in the near future. And also, This research progressed from the results of previous study that the forecasting accuracy of model could be changed according to data apply to forecasting model. The models to evaluate consisted of exponential smoothing model, ARIMA model and regression model from among the models confirmed in previous studies on accuracy. The data for analysis has applied the number of Japanese tourist in 15 years. Each model was assessed based on MAPE(mean absolute percentage error) and the probability of improvement of forecasting accuracy was additionally evaluated by combined techniques. The results of model estimation indicates that Winter's exponential smoothing model was measured as the best model(MAPE=8.72%) in individual model, and also the combination model consisted of Winters exponential smoothing model and cubic regression model was relatively outperformed(MAPE=8.63%) the other combined techniques.
