The hospitality industry’s dynamic nature demands accurate forecasting of hotel room prices to optimize revenue management strategies. This paper presents an experimental study assessing machine learning techniques and time series analysis for forecasting hotel room prices. We enhance prediction accuracy by leveraging historical booking data, seasonal patterns, and hotel characteristics. We employ time series models, including AutoRegressors and Prophet, to capture underlying trends and seasonal variations. We also evaluate machine learning models such as random forest, gradient boosting machine, extra trees regressor, and neural networks. These models are trained on features like booking lead time, historical hotel occupancy, room nights, and number of adults to capture complex relationships influencing prices. Our methodology is demonstrated through a case study using over 40,000 reservations from a Brazilian hotel over a decade. Experimental results show that tree-based models performed best, with the Gradient Boosting achieving 6.94% of NRMSE and 31.26 of MAE. Our findings contribute valuable insights for price estimation in the hospitality industry, offering robust methods to enhance revenue management strategies.