Quant Convergence: Bridging Classical Value Investing and Modern Factor Models

This research tests whether Benjamin Graham's classic value investing rules can act as a mathematical filter to prevent complex machine learning models from memorizing market noise. The study compares pure Graham rules, modern factors, and a combination of both against XGBoost and AutoGluon models using 20 years of S&P 500 data.

• The AutoGluon model achieved 222.68% returns but suffered a 39.78% maximum drawdown by buying volatile tech stocks before the crash.
• The pure Graham Random Forest yielded the highest overall return of 232.13% with a low risk profile, indicated by a 1.38 Calmar Ratio.
• The Combined Random Forest generated 202.91% returns while maintaining the lowest maximum drawdown of 34.53% among all tested models.

The findings demonstrate that Graham's "margin of safety" remains an effective method for controlling risk in modern AI-driven equity selection.