GASPRO International Journal of Eminent Scholars

Guides For Authors

Call For Papers
Processing Charges
Journal Coverage
Open Access Policy
Terms and Policies

Secured Payment

INTEGRATING MACHINE LEARNING INTO ASSET PRICING MODELS: ADVANCING FORECAST ACCURACY AND FINANCIAL STABILITY IN U.S. MARKETS

ABSTRACT

The study examined the use of machine learning (ML) techniques in classical asset pricing models for the improvement of return prediction and risk management in the U.S. financial markets. Some widely used models, including the Capital Asset Pricing Model (CAPM) (LINTNER 1965) and the Fama French multifactor model, are unable to fully describe the complex, nonlinear dependence of variability in financial data and trading activities. ML models – Random Forest, Gradient Boosting, and Neural Networks – are used to AdaBoost the hybrid models on the U.S. stock market data. The relationship was also captured with the use of a graph. The data used ranges from 2015 to 2025. The superior reliability of ML models obtained in earlier studies provides the basis for the finding that ML can significantly enhance out-of-sample return forecasts. This capability also has the potential to increase market stability by empowering investors and policymakers to anticipate falls and adjust

accordingly.

KEYWORDS: Integrating, Machine, Learning, Pricing, Models, Forecast, Accuracy, Financial Stability, U.S Markets.

Olatunji Ahmed. A

INTERCONTINENTAL JOURNAL OF EDUCATION, SCIENCE AND TECHNOLOGY VOL 9. NO. 1