The Number of Occupied Hotel Rooms: A Time Series Model that Accounts for Constrained Capacity and Prices

The daily number of occupied hotel rooms in three large Swedish cities is modelled by an integer-valued and binomial autoregression. The model includes the capacity constraint and price variables are incorporated through the parameters of the model. The model implies a duration of hotel visit and an occupancy probability. We find that a 10 percent increase in the price level shortens the median duration of a hotel visit by approximately 6 percent during weekends and 8 percent during weekdays.