sachahuberty/Portfolio-Optimization-Machine-Learning

Cross-sectional stock selection on the S&P 500 using fundamental, technical, and macro factors. XGBoost classifies direction while meta-labeling sizes positions via triple-barrier labels. Purged cross-validation prevents lookahead bias, sector/beta neutralization controls risk, and cost-aware backtesting evaluates the resulting long-short portfolio

语言：Jupyter Notebook · GitHub 星标：16 · Fork：2

在 GitHub 上查看原仓库