Forecasting Hotel Occupancy Rates with Time Series Models: An Empirical Analysis

Abstract

This study examines empirically the use of two time series models, Box-Jenkins and exponential smoothing, for forecasting hotel occupancy rates. The models are fitted and tested using actual monthly occupancy rates for a major center-city hotel. Both models show a high level of predictive accuracy. Since these models are relatively easy to implement, they should be very useful in actual hotel operations and other applications such as yield management.