Designing a Diversified Automated Futures Trading System
Summary
The article describes the components and operating choices of a long-running automated futures system. Its rules combine trend and momentum signals, breakouts, carry, relative signals, skew, acceleration, and mean reversion. Forecasts are weighted across rule families, attenuated when volatility is elevated, and adjusted for instrument-specific trading costs. Portfolio exposure is also allocated across asset classes, with dynamic optimisation used in the broader system.
The author explains how markets enter the research universe: survey broker offerings, collect historical data, sample prices, and only then enable eligible instruments for trading. Duplicate, unreliable, legally restricted, illiquid, or costly markets may be excluded. Reported rule Sharpe ratios are described as crude because they equally weight instruments, including markets too expensive to trade those rules. The article includes backtest statistics, but these are not live results and the provided text gives limited detail on validation, implementation assumptions, or the mechanics of several sections. The system is presented as one practitioner's configuration rather than a universal recipe.
Key ideas
- The system combines multiple futures rules, including trend, momentum, breakout, carry, skew, and mean reversion.
- Markets are sampled and assessed before they are added to the tradable configuration.
- Trading eligibility depends on legal access, liquidity, and estimated costs.
- Forecast weights and volatility adjustments vary across rules and instruments.
- The reported rule performance is crude because it equally weights instruments regardless of their trading costs.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.