Adaptive Cross-Sectional Factor Trading for Perpetual Futures
Summary
This system describes an adaptive long/short strategy for Binance USDⓈ-M perpetual contracts. It starts from five seed factors spanning momentum, reversal, funding, premium, and open interest, then evaluates additional candidates through a constrained factor factory. Cross-sectional factor values are clipped at their tails and converted to rank scores; a composite combines eligible factor scores using adaptive weights. The design also includes a universe, BTC-beta hedging, turnover and drawdown controls, factor lifecycle rules, and forward shadow evaluation.
Execution is treated as part of the strategy: a persistent state machine reconciles uncertain order outcomes, and the system offers shadow, paper, and live modes. The description and partial source give substantial architectural detail but no reported return, risk, or comparative test results. Consequently, they explain how the framework is intended to operate, rather than establishing that its factor selection, adaptive weighting, or execution design produces profitable live performance.
Key ideas
- The system combines cross-sectional momentum, reversal, funding, premium, and open-interest factors.
- Candidate factors pass through constrained generation and forward shadow evaluation.
- Factor scores are ranked across the eligible universe and combined with adaptive weights.
- Portfolio and execution controls include beta hedging, turnover limits, drawdown controls, and order reconciliation.
- The document reports system design but provides no strategy performance evidence.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.