Precipitation within specific return periods plays a crucial role in the design of hydraulic infrastructure for water management. Traditional analytical approaches involve collecting annual maximum precipitation data from a station followed by the application of statistical probability distributions, and selecting the best-fit distribution based on goodness-of-fit tests (e.g., Kolmogorov-Smirnov). However, this methodology relies on current data, raising concerns about its suitability for outdated data. This study aims to compare Probability Density Functions (PDFs) with the Random Forest (RF) machine learning algorithm for estimating precipitation at different return periods. Using data from five stations located in various parts of the Arequipa province in Peru, it was evaluated the performance of both methods using the Root Mean Square Error (RMSE) metric. The results show that RF outperforms PDFs in most cases, yielding lower RMSE values for precipitation estimates at return periods of 2, 5, 10, 20, 50, and 100 years for the studied stations.