Assessing Hotel Reservation Forecasting Models to Inform Dynamic Pricing Strategies
Assessing Hotel Reservation Forecasting Models to Inform Dynamic Pricing Strategies

Author(s): Majlinda Godolja, Arjola Sinani
Subject(s): Social Sciences
Published by: Udruženje ekonomista i menadžera Balkana
Keywords: Hotel reservation forecasting; Dynamic pricing; SARIMA model; Prophet model; Monte Carlo Simulation; Forecast Accuracy Metrics
Summary/Abstract: Accurate forecasting of hotel bookings is crucial for optimizing dynamic pricing strategies and ensuring efficient resource management in the hospitality industry. This study analyses daily booking data from a hotel from Vlora, Albania, covering two distinct periods: June 2021 - April 2022 and May 2022 - January 2023. It evaluates the performance of two widely used forecasting models: SARIMA, known for its robustness in capturing seasonal patterns, and Prophet, recognized for its flexibility in handling irregularities and holiday effects. Forecast accuracy and stability are analyzed using key performance metrics (MAE, MSE, RMSE, R²), with Monte Carlo simulations providing additional insights into forecast variability under different scenarios and enhancing the reliability assessment of each model. The findings reveal that model performance varies across different periods, highlighting the need for a dynamic model selection approach that adapts to evolving booking trends. These insights offer practical guidance for hotel managers, supporting improved reservation management and data-driven pricing strategies to enhance overall operational efficiency.

• Details
• Contents