Effective Machine Learning Methods Used for Stock Price Forecasting: A Performance Evaluation
A crucial component of financial decision-making and investment planning is stock price forecasting, but it is a complicated problem because of the non-linear and unstable dynamics of the market. In this paper, the researcher outlines a functioning stock market forecasting model which aims at enhancing the degree of forecasting precision founded on the past data of the S and P500 index. The approach involves the use of systematic data gathering and a thorough preprocessing procedure, such as the management of missing data, feature scaling with standardization and chronologically dividing the dataset into training (80%) and testing (20%) sets. Machine learning (ML) models called Random Forest (RF) and Linear Regression (LR) are used to analyze current market trends and forecast future prices. The key error measures (R2, MSE, RMSE and MAE) are used to evaluate the performance of each model, which guarantees an accurate evaluation of forecasting reliability. The results of the experiments prove that both models provide very accurate predictions, though LR has slightly better error performance, and RF has high levels of generalization. The comparative analysis also demonstrates that proposed models are superior to the current models, such as ANN and SVM. Overall, the work offers an informative, evidence-based prediction model, which is resistant, stable, and can be used in supporting real-time financial decisions and investment policies. Keywords: Stock Price Forecasting, S&P 500 Index, Machine Learning, Financial Market Prediction, Linear Regression, Random Forest, Stock Market Analysis.
Authors
- Sridevi Allu
- Pradeep Rao Vennamaneni (ORCID: https://orcid.org/0009-0006-7994-3918)
- Khaja Moiezuddin Mohammed
- Rashmika Jalagam
- Venkata Phaneendra Reddi
- Sri Prudhvi Adusumilli
Institutions
- American Airlines (United States) (US)
- IQVIA (United States) (US)
- Charter Communications (United States) (US)
- Atlantic Health System (US)
Publication Details
- Journal
- American Journal of Computing and Engineering
- Published
- 2026-11-22
- DOI
- https://doi.org/10.47672/ajce.3036
- Primary Topic
- Stock Market Forecasting Methods
- Type
- article
- Field-Weighted Citation Impact
- 0.00