Measuring and Combining Crypto Perpetual Futures Alpha Signals
Summary
The article presents a workflow for studying and combining signals on Binance crypto perpetual futures. It examines carry from funding rates and cross-sectional momentum alongside a breakout measure based on closeness to recent highs. The author first restricts the asset universe by trailing dollar volume, then inspects feature distributions, scales carry and momentum with daily z-scores or decile ranks, and evaluates their relationship with forward returns. Signal strength, decay, stability, noise, and correlation are framed as inputs to portfolio weighting decisions.
The example combines cross-sectional carry and momentum with a time-series breakout overlay, which can tilt the portfolio net long or short. The article shows illustrative portfolio analyses and notes that delaying execution by a day changes the return calculation. Its evidence is exploratory rather than conclusive: signal diagnostics are described as shallow, and the stated feature alignment can assume execution at the same close used to compute the features. It does not establish robustness after turnover, costs, or more careful universe selection; those questions are deferred to later work.
Key ideas
- Assess each signal’s scaling needs, predictive strength, decay, stability, and correlation before combining it.
- The example uses funding carry and momentum as cross-sectional predictors and recent-high breakout as a time-series overlay.
- Daily z-scores and decile ranks are used to scale heavy-tailed carry and momentum features.
- A time-series overlay can make a long-short portfolio net long or net short.
- Feature-return alignment, delayed execution, turnover, transaction costs, and signal stability affect how credible portfolio results are.
Tags
This summary was written by Stratmill's research agent from the original; it is not a copy of the source.