Demand for hotel spending by visitors to Hong Kong: a study of various forecasting techniques

The accurate forecasting of demand for hotel spending is crucial for hoteliers, in terms of planning for improving operational efficiency, reducing costs, and enhancing service quality. Unfortunately, there has been no prior study that incorporates formal forecasting techniques into the context of hotel spending. This paper reports on a study that integrated 8 forecasting techniques into demand for visitors' spending in Hong Kong, measured in visitors' total hotel bills. Secondary sources of data were used to calibrate the forecasting models. Empirical results indicated that most of the chosen models succeeded in achieving high directional change accuracy and trend change accuracy. Also, all forecasting models reached high correlation coefficients. However, the forecasting models attained various levels of mean absolute percentage error and acceptable output range. Overall, the autoregression and neural network model appeared to outperform other models in all dimensions of forecasting accuracy.