Domain expertise first
Our edge starts with direct experience of the physical commodity markets we trade. The team has worked in physical trading, agronomy, statistics and systematic portfolio management. That experience shapes which questions we ask and which data we trust.
Data and models
Proprietary machine learning models take in a broad range of fundamental and alternative datasets. We combine their signals with quantitative research and fundamental trading expertise.
Models are hypothesis-driven, so each signal should be grounded in observable economic behaviour. Researchers build, test and interpret every signal before it is used.
Market-neutral portfolio construction
The Flagship Commodity Strategy builds a market-neutral portfolio across more than 100 global commodity futures markets in energy, metals and agriculture. Long and short positions remove directional beta exposure, so results depend on differences between markets rather than the direction of commodity prices as a whole.
Portfolio construction is fully interpretable. It never depends on opaque or black-box systems.
Risk
- Portfolios are stress tested against historical crises and against simulated regimes that markets have not yet produced.
- Clustering and machine learning analytics monitor factor exposure, concentration and regime risk.
- The framework is built to adapt as market regimes change.
Learn more
See our strategies, meet the team, read the FAQ or contact us. Information on this site is for qualified investors and is not an offer to sell or a solicitation to buy any security.
Talk to us
Book a 30-minute call with our team.
For professional and qualified investors only. Nothing on this site is an offer or solicitation.