Research · One Position, Four Bets
Managing Equity Risk via Hierarchical Orthogonal Decomposition
The series argues that style labels and broad-market beta are too coarse for modern portfolio review. A single stock position can be decomposed into market, sector, subsector, and residual bets, with each layer measured after the prior layer is removed.
Series library
Parts 1–2 mirrored locallyOrthogonal Decomposition
Stripping market and sector noise to isolate subsector risk.
Mirrored article
One Position, Four Bets
Part 1 is available as a local mirror for archival and SEO purposes, with the Medium link preserved. The full article route renders the canonical markdown and generated figures from the research pipeline.
Article thesis
Same ticker label, different economic bet: market, sector, subsector, and residual components have to be separated before a position can be called diversified.
Medium mirror: https://medium.com/@ConradGann/3f29c180fc79
Mirrored article
Risk Structure in 13F Filings
Part 2 applies the same four-layer decomposition to five concentrated 13F filers — portfolio risk structure, active dollars, concentration, turnover, and what survives realistic filing lag.
Article thesis
Allocator diligence compresses active books into summary statistics. The framework separates market risk, thematic positioning, and stock-specific selection — and shows which of that structure persists when holdings are observed with a 45-day lag.
Medium mirror: https://medium.com/p/6883b93ee10f
Series structure
Part 1 — The problem. Custodial reporting obscures concentration. Four names that looked diversified drew down 50%+ together in 2022. Style factors (Growth, Value) are symptoms, not drivers — subsectors are the real unit of risk.
Part 2 — The manager. The same decomposition applied to five concentrated 13F filers: how portfolio risk partitions across market, thematic, and stock-specific layers; how active structure compounds in dollars; and what survives a realistic filing lag.
Part 3 — The attribution standard. Hierarchical orthogonalized regression: a three-level cascade (Market → Sector → Subsector) that strips embedded exposures and produces clean, additive variance attribution. No double-counting, no multicollinearity, no latent factors.
RiskModels ecosystem
Research here. Reproduce through the API. Operate in the web app.
RiskModels.org stays the credibility layer: methodology, proof, and exhibits. Product links are kept contextual so the research remains the primary object.
Research
RiskModels.org
Methodology, article series, and public exhibits for institutional review.
Read the researchAPI
riskmodels.app
REST API, SDKs, CLI, and MCP-ready endpoints for reproducible decomposition calls.
Open API docsDashboard
riskmodels.net
Web application surface for portfolio workflows, dashboards, and authenticated product use.
Open web app