Modelling and prediction of a destination’s monthly average daily rate and occupancy rate based on hotel room prices offered online

Abstract

Tourism metrics are essential for managing a destination. Hotel performance metrics such as average daily rate and occupancy rate are two of the most prominent metrics for the industry. The authors’ research group works on developing methods for estimating tourism metrics based on digital footprint. Data available publicly on the Internet, including hotel room prices, are collected daily. This article shows that the prices offered online have a high positive correlation with those reported by official statistics at the Nomenclature of Units for Territorial Statistics 2 level after the online prices have been preprocessed and, thus, the relevance of this data source is established. This article then presents a model for explaining and predicting mean hotel occupancy rates by destination based on these prices. The results are very promising, the fit is excellent and the predictions are also good. In summary, prices have moved from reflecting the expected demand to reflecting the actual demand and occupancy rate.