A New Approach to Modelling and Forecasting Monthly Guest Nights in Hotels

Starting from a day-to-day model on hotel specific guest nights we obtain an integer-valued moving average model by cross-sectional and temporal aggregation. The two parameters of the aggregate model reflect the daily mean check-in and the daily check-out probability. Letting the parameters be functions of dummy and economic variables we demonstrate the potential of the approach in terms of interesting interpretations. Empirical results are presented for a series of Norwegian guests in Swedish hotels. The results indicate strong seasonal patterns in both mean check-in and in the check-out probability. Models based on differenced series are preferred in terms of goodness-of-fit. In a forecast comparison the improvements due to economic variables is minute.