Year: 2025
Type: article
Abstract: This study explores the application of a Random Forest Classifier to predict stock performance among S&P 500 companies based on financial and stock performance data. Financial indicators such as Return on Equity (ROE), Return on Assets (ROA), Market Value (MV), Price-to-Book Ratio (PB), and Price/Earnings-to-Growth Ratio (PEG) were combined with stock performance data collected between September 2023 and June 2024. The data was merged into a unified dataset and split into training and testing sets. The model was trained using financial features from September 2023 to March 2024, while performance predictions were tested on data from March 2024 to June 2024. The Random Forest model achieved an accuracy of 63.92%, highlighting its effectiveness in identifying unsatisfied stocks but showing moderate accuracy for higher performance categories. The findings underscore the model's potential for data-driven stock selection, while also suggesting that further improvements, such as feature selection optimization and additional data integration, could enhance prediction accuracy.
Authors: Chang Liu, Chengyongyi Zhan, Yue Zhang, Yuqi Cai
Language: English
Field: Decision Sciences
Domain: Social Sciences
Open Access status: hybrid
OpenAlex ID: W4416396609