Comparative study of the Box-Jenkins and Holt-Winters model in forecasting tourist arrivals / Nur Shafiqah Jumery, Fatin Amira Ahmad Termizi and Nurul Atiqa Mulyadi

Tourism can be described as the activities of persons known as visitors. A visitor is referring to someone who makes a visit to a main destination outside their usual environment for less than a year for any main purpose which including holidays, leisure and recreation, business, health, education or other purposes. Malaysia has many public holidays that causes number of tourist arrivals to Malaysia constantly changing every year. In addition, one of the sources of country's income contributed from the tourism sector. Thus, it is important to forecast the number of tourist arrivals. Hence, the forecasting beneficial for tourism agencies to improve their marketing techniques to attract more tourists to come to Malaysia. Furthermore, this paper is attempt to compare the accuracy of the techniques between Box Jenkins and Holt-Winters model. The sample data for this study were obtained from Tourism Malaysia Department, which consists of number of tourist arrivals against three countries which are Singapore, Korea and United Kingdom (UK) from year 2013 until year 2017. Microsoft Excel was used to examine the data for Holt-Winters meanwhile E-Views 7 was used to find the best Box Jenkins model. By comparing the value of Mean Square Error (MSE), Root Mean Square Error (RMSE), and Mean Absolute Percentage Error (MAPE), the findings of this study suggest that besides Box Jenkins, Holt-Winters model also suitable to be used in forecasting tourist arrivals. Hence, the Holt-Winters model are the appropriate model to forecast the tourist arrivals from United Kingdom (UK) and Korea. While, the SARIMA (1,1,1)(1,1,1)12 is the appropriate model for forecasting tourist arrivals from Singapore. Therefore, Holt-Winters and Box Jenkins model are suitable used to forecast and determine the future trend of tourist arrivals. Last but not least, it is crucial to make an improvement in the accuracy of forecasting techniques and any further research regarding the tourist arrivals forecasting would be encouraged.