Single or Combine? Tourism Demand Volatility Forecasting with Exponential Weighting and Smooth Transition Combining Methods

Tourism forecasting has garnered considerable interest. However, integrating tourism with volatility is significantly less typical. This study investigates the performance of both single models and their combinations for demand. The seasonal autoregressive integrated moving average (SARIMA) model used to construct mean equation, three models, namely generalized conditional heteroscedasticity (GARCH) family error-trend-seasonal exponential smoothing (ETS-ES) model, innovative smooth transition (STES) are employed estimate monthly tourist arrivals into Malaysia. also assesses accuracy forecasts using simple (SA), minimum variance (MV), novel (ST). STES performs best out-of-sample demand volatility, followed closely by ETS-ES. In contrast, ST combining method surpasses SA MV. Interestingly, forecast methods do not always outperform but they consistently worst model. MCS DM tests confirm aforementioned findings. article merits consideration future research on volatility.