Philosophy

What We Believe About Portfolio Construction

Conviction-driven, mathematically grounded portfolio construction. Built for people who want to understand their portfolios — not just own them.

01

Structure Over Guesswork

Portfolios should be engineered, not assembled from marketing narratives. Every allocation in our platform has a mathematical basis — an optimization objective, a constraint set, and a traceable decision chain. There is no "trust us" in the output. You see the logic.

02

Evidence Over Narrative

Every strategy claim is backed by walk-forward backtesting against real market data. No hypothetical returns. No cherry-picked timeframes. No survivorship bias. Transaction costs are modeled. Results are deterministic and fully reproducible.

03

Transparency Over Black Boxes

See the full decision chain: from covariance estimation through constraint binding to final weight assignment. Every optimization generates a 17-file artifact bundle — weights, risk decomposition, attribution, constraint reports, and optimizer diagnostics. Nothing is hidden.

Academic Foundations

Every engine in the platform is grounded in peer-reviewed financial mathematics. These are the papers and methods that form the theoretical backbone.

Markowitz (1952)

Modern Portfolio Theory

The foundation of all portfolio optimization. Risk-return trade-off, efficient frontier, the case for diversification over stock-picking.

Ledoit & Wolf (2004)

Oracle Approximating Shrinkage

Covariance matrix shrinkage that makes high-dimensional estimation reliable. Powers the covariance input of our Robust Mean-Variance engine — the only return-aware optimizer we ship.

López de Prado (2016)

Hierarchical Risk Parity

Clustering-based allocation that works when Markowitz fails. Immune to covariance matrix instability. Powers our Smart Diversification strategy.

Rockafellar & Uryasev (2000)

CVaR Optimization

Linear programming reformulation for minimizing expected tail losses. Powers our Downside Guard strategy — the losses that matter most.

Black & Litterman (1992)

Bayesian Return Integration

Combines market equilibrium returns with investor views in a Bayesian framework. Produces intuitive portfolios from subjective inputs.

Spinu (2013)

Cyclical Coordinate Descent

Efficient solver for the Equal Risk Contribution problem. Powers our Risk Parity and Risk Budgeting engines.

Not a Broker. Not a Robo-Advisor.

Holy Funds is a portfolio intelligence platform. We don't automate your decisions — we give you the tools, data, and structure to make better ones. Research strategies. Test allocations. Understand risk. Build with confidence.