Comparison between multiplicative Holt Winter and decomposition method in predicting the number of incoming international tourists to Indonesia

3rd Indonesian Operations Research Association - International Conference on Operations Research 2018

计算机科学

Supriatna, A.^1 ; Lesmana, E.^1 ; Aridin, L.^1 ; Sukono^1 ; Napitupulu, H.^1

Department of Mathematics, Faculty of Mathematics and Natural Sciences, Universitas Padjadjaran, Bandung, Indonesia^1

关键词: Data patterns;    Decomposition methods;    Error levels;    Forecast accuracy;    Forecasting methods;    Mean absolute percentage error;    Policy making;    Tourism sectors;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/567/1/012047/pdf
DOI  :  10.1088/1757-899X/567/1/012047

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Indonesia has a remarkable potential in tourism sector, hence government should regulate all matters related with tourism and pay more attention to the developments happening in the tourism sector, including policy-making and planning. One way to aid the government in policy making and planning is by conducting a prediction about the number of incoming international tourists to Indonesia. This paper will elaborate the analysis about the comparison between multiplicative Holt-Winter and decomposition method in predicting the number of incoming international tourists to Indonesia. This paper aims to determine the appropriate forecasting method suitable for the available data pattern and provides more accurate forecasting result. Forecasting could be conducted using multiplicative Holt-Winter method and decomposition method. Meanwhile to determine the error level and forecast accuracy, mean absolute percentage error (MAPE) is used. Analysis result showed that prediction using Holt-Winter method provides MAPE of 4.7098%, meanwhile the MAPE for decomposition method is 3.3254%. It could be concluded that to determine the number of incoming international tourist to Indonesia, decomposition method is more appropriate.