Comparing HRP and HERC in a Cross-Sectional Momentum Portfolio

Summary

This project outlines a market-neutral long-short portfolio framework that selects assets with cross-sectional momentum and compares equal weighting, inverse-volatility weighting, Hierarchical Risk Parity (HRP), and Hierarchical Equal Risk Contribution (HERC). It describes a feature pipeline built from technical and volatility indicators, a train/test data split, and periodic rebalancing across a broad CFD universe. Candidate momentum settings vary lookback, skipped recent data, and holding period; allocation methods then assign normalized unlevered weights.

A vectorized backtest compares thousands of combinations, and Monte Carlo simulation is proposed for terminal wealth and maximum drawdown when sizing capital. The reported top performers are mainly HRP and HERC, with inverse volatility also appearing among the stronger methods, but the best Sharpe ratio is only 0.5 and the author says the system is not ready to trade. Claims of low overfitting risk are tentative: the study uses a single train/test path, lacks a representative benchmark, omits beta considerations, and leaves transaction costs, slippage, leverage, and stronger validation as future work.

Key ideas

• Cross-sectional momentum ranks assets for long and short selection during periodic rebalancing.
• The framework compares equal weighting, inverse volatility, HRP, and HERC allocations.
• Technical indicators are engineered as features, with non-stationary features removed in the described process.
• Backtests report HRP and HERC among the strongest allocation approaches, but the reported best Sharpe ratio is modest.
• A single train/test split and omitted costs, benchmark, and beta limit conclusions about robustness and tradability.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.