A Study on Sailboat Price Prediction Method Based on Multiple Linear Regression Model and XGBoost Algorithm

this study aims to explore sailboat price prediction methods through the utilization of a multiple linear regression model and the XGBoost algorithm, addressing the pricing challenges in the second-hand sailboat market. The flourishing global economy has led to widespread interest in the second-hand vessel market. This paper provides a dataset containing detailed information on approximately 3,500 sailboats, encompassing attributes such as brand, length, geographic region, and listing price. The model is built upon sound assumptions and constraints, ensuring the reliability of the outcomes. Data preprocessing involves merging, handling outliers, addressing duplicates, and reducing data dimensions. PCA and stepwise regression are employed for dimensionality reduction. The model employs both the multiple linear regression and XGBoost algorithm to establish predictive models. Through a case study involving the Hong Kong market, the model's predictive performance is optimized. Results demonstrate that the model exhibits strong predictive capabilities for specific regional markets. Through the integration of the multiple linear regression model and the XGBoost algorithm, this study presents a valuable sailboat price prediction approach, offering decision support for market practitioners and investors. The research outcomes hold practical applications and academic value in the context of second-hand sailboat transactions.