Applying SARIMA, Grey Prediction and BNP models to forecast the tourism demand from Mainland to Taiwan.
Date Issued
2016
Date
2016
Author(s)
TU, YU-YIN
Abstract
There has been a growing interest in tourism demand forecasting over the past decades because of the constant growth of world tourism. Although there has the consensus on the need to develop more accurate forecasts and the recognition of their corresponding benefits, there is no one model that stands out in terms of forecasting accuracy. Since 2008, the amount of Mainland tourists to visit Taiwan has been growing rapidly. This study aims to build SARIMA, seasonal GM(1,1) model and Back Propagation Neural Network model to forecast the tourism demand from Mainland to Taiwan. This study chose the period from January 2009 to December 2015, 72 monthly data, to build SARIMA model, seasonal GM(1,1) model and seasonal Back Propagation Neural Network model, and evaluates the prediction by MAPE and RMSPE. Via comparing MAPE and RMSPE, we select seasonal Back Propagation Neural Network model which has the lowest MAPE to forecast 2016 and 2017 Mainland tourists to Taiwan.
Subjects
SARIMA model
seasonal GM(1,1) model
BNP model
tourism demand
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-105-R01341015-1.pdf
Size
23.54 KB
Format
Adobe PDF
Checksum
(MD5):b9e7d220319fc908cc5be4c21cceb315