Airbnb pricing in Los Angeles: insights from exploratory analysis and predictive modeling

As Airbnb and other short-term rental platforms continue to transform the hospitality industry, understanding the factors that influence listing prices is increasingly important for hosts, guests, and platforms. This study analyzes a large dataset of Airbnb listings in the greater Los Angeles region to examine pricing determinants using exploratory data analysis and predictive modeling. The dataset includes 55 variables capturing listing attributes, amenities, host characteristics, policies, neighborhood features, and review metrics. Feature selection techniques including Pearson correlation, Chi-square test, and Recursive Feature Elimination with Random Forest are applied to identify the most informative predictors. Multiple regression and classification models are then evaluated with systematic hyperparameter tuning to assess predictive performance and robustness. Interpretability analysis using SHAP (SHapley Additive exPlanations) on the best-performing model further reveals how key features contribute to pricing variation. The findings provide actionable insights and recommendations for hosts seeking to optimize pricing strategies and contribute to a deeper understanding of pricing dynamics in the sharing economy.