Tourist Arrivals in Malaysia Post-COVID-19: A Comparison Between Holt-Winters and ARIMA Model

Tourism is a major contributor to Malaysia's socioeconomic growth, with visitors' investments contributing to the country's Gross Domestic Product (GDP). To capitalise on this sector, the government has announced the initiation of the Visit Malaysia 2020 campaign. However, in December 2019, the globe was threatened by the COVID-19 virus, and the first case was quickly identified in Malaysia. The number of cases grew over time, and on 11 March 2020, the World Health Organization (WHO) proclaimed the COVID-19 outbreak a pandemic. COVID-19 has caused substantial damage and severely affected the country's economy. To ensure the economy can return to normal, a systematic recovery plan must be adopted promptly. Therefore, the focus of this study is to examine the pattern of tourist arrivals before and after the COVID-19 pandemic to forecast tourist arrivals in Malaysia using the Autoregressive Integrated Moving average (ARIMA) and Holt-Winters models. The data on tourist arrivals in Malaysia from January 2018 until June 2022 were obtained from the Tourism Malaysia website. The results of this study reveal that the ARIMA model outperforms the Holt-Winters model, with the ARIMA (1,1,1) model providing the best fit for forecasting tourist arrivals in Malaysia, characterised by the lowest values of Mean Squared Error and Mean Absolute Percent Error. Future research might be conducted on alternative methods linked to the ARIMA model, such as the Fuzzy Seasonal ARIMA model (FSARIMA), or on distinct types of data series.