Short-Term Forecasting of Tourist Arrivals to Sri Lanka from Asian Region

Sri Lankan tourism market consists all regions of the world. The Asian region is the highest tourist producer to Sri Lankan market. Short term forecasting plays a vital role in the process of operational level decision making in the tourism industry. It is a known fact that the expectations of the tourists are not same for all the regions. As such it is necessary to forecast the number of arrivals region wise, in order to satisfy the expectations of the customers to obtain the optimum benefits to the country. On the view of the above, the study was focused on identifying suitable short-term forecasting techniques to arrivals from Asian region to Sri Lanka. Monthly tourist arrivals data from Asian region for the period of January 2008 to December 2014 was obtained from Sri Lanka Tourism Development Authority (SLTDA). Moving Average (MA) Smoothing Models (SM), Single Exponential Smoothing (SES) Models, Double Exponential Smoothing (DES) Models and Holt's Winters' Three-Parameter Models (HWTPM) were tested during in the study. Auto-Correlation Function (ACFs), Residual plots, and Anderson-Darling test were used to test the model validation criterion, while the Mean Absolute Percentage Errors (MAPE) were was used to select the best-fitting model. The results of the study revealed that the residuals of DESM model with α = 0.72, γ = 0.72 are normally distributed and uncorrelated. MAPE of the model is 1.5%. However, smoothing techniques can be accurately used only for short-term forecasting. The results of this study facilitate for short-term decision-making in many operations within the tourism industry. It is recommended to test non-linear models and circular model for more accurate forecasting. Further, it is better to concern forecasting Western and Eastern Europe as well