Combining time-series and econometric forecast of tourism activity

It has always been difficult to model the travel industry because tourism involves such a diverse set of activities. However, various regional decision makers have become increasingly interested in predicting the flows of visitors through their market. Accurate forecasts of the number of tourists' arrivals, their length of stay, and their expenditures improve planning and inventory control. Stochastic time-series models have compared favorably with econometric models at the aggregate level while some naive automatic forecasting tools have fared well in comparison when predicting industry-level behavior. Several approaches have been developed to improve forecast accuracy. This paper presents parsimonious methods of improving accuracy by combining various forecasting techniques. The Box-Jenkins stochastic time-series method is combined with a traditional econometric technique to forecast airline visitors to the State of Florida.