Track 04Closed

Spread Trading — Calendar / Basis / Cross-Asset

trade perp-vs-quarterly spreads, the basis premium, and cointegrated pairs

Key metric

MMs hold the spread <1 bps; no pair is stably cointegrated

Why an edge should exist / hypothesis
Classic statistical arbitrage of spreads (the de Prado spread approach).
Construction
Calendar (perp vs quarterly), dynamic basis, cross-asset cointegration.
Test data
Multi-exchange, 12 months.
Results
Calendar: funding=basis, quarterly liquidity ~0.1% of perps. Dynamic basis: MMs hold <1 bps. Cross-asset: BTC/ETH cointegrated only 33% of windows.
Validation
All 4 types negative/untradable.
Root cause
The perpetual-funding mechanism makes the de Prado spread approach inapplicable to crypto; quarterly contracts are illiquid.
Verdict
Dead (all 4 types). Basis is conditional under sustained contango.
Source
archive/SPREAD_TRADING_RESEARCH.md

Want the same check on your strategy?

Validation →