| Abstract: | This research uses a big dataset of online prices published on Booking.com by three-, four- and five-star hotels located in Madrid (Spain). Data is broadened by other sources, resulting in a rich set of context-, hotel- and offer-based variables. This research aims to determine the impact of these variables on the online room prices set by one representative sample, featuring the total pricing behaviour as well as per hotel categories. The variables considered, their extent and the insight per category represent a novelty and complement the literature on demand forecasting and hedonic pricing, enabling the improvement of optimisation techniques. The models, based on regression analysis with random effects, reveal a significant impact of the variables on room prices and a clearly distinct behaviour for five-star hotels. This research contributes to the scientific community and practitioners by showing the pricing dynamics behaviour of an important urban destination by star rating. |
