The risk model

One model,
four levels of risk.

Sectors, sub-sectors, styles and trading factors, and idiosyncratic risk for each future. Cross-sectional decomposition across more than 40 commodities.

Model structure

The ARC nested commodity factor modelThree sectors resolve into eleven sub-sectors and are surrounded by seven cross-sectional style and trading factors.BASISMOMENTUMSHORT MOM.OPEN INTERESTACTIVITYVOLATILITYSHORT VOL.PRECIOUSBASECRUDENAT GASREFINEDBIOFUELSCOALGRAINSSOFTSPROTEINSLUMBERMETALSENERGYAGSARC
01Sectors3
02Sub-sectors11
03Styles & trading factors7
04Idiosyncratic, each future1,200+

Level 03 — styles & trading factors

  • Basis
  • Momentum
  • Short-term momentum
  • Open interest
  • Trading activity
  • Volatility
  • Short-term volatility

Research applications

One model,
two research applications.

ARC studies historical return patterns in style factors and relationships between contracts with comparable modeled exposures. Both applications are estimated from the same factor model.

Designed to support a portfolio manager's own alpha research, portfolio construction and risk process—not to replace investment judgment.

Research framework

One model, two research applicationsThe ARC factor model supports historical style-factor analysis and relative-value research at the pair level.ONE FACTOR MODELSTYLE FACTORSLONG AND SHORTPAIR LEVELRELATIVE VALUEHISTORICAL CORRELATIONρ ≈ 0

Two applications / one estimation

A nesting commodity factor model built for institutional research workflows

From model to workflow

Use the model without changing how you work.

ARC supplies analytical outputs as research-ready flat files. Start with a representative portfolio or sample data, then determine whether the model adds information to your existing process.

Portfolio research

See what sector arithmetic misses.

A portfolio can appear balanced across Energy, Metals and Agriculture while carrying material basis, momentum or volatility exposure.

View the exposure fingerprint →

Analytical data

Flat files, transparent inputs.

Factor exposures, factor returns, covariance estimates and idiosyncratic returns can enter existing research, backtesting and portfolio systems.

Review the methodology →

Evaluation

Begin with your own use case.

Request sample data or ask ARC to discuss how the model would read a representative commodity portfolio.

Contact ARC