Cointegration
Cointegration describes a state where two or more time series move together over the long term despite individual short-term fluctuations. In crypto markets, cointegrated assets share a common stochastic trend, meaning their price spread remains bounded.
Unlike correlation, which measures linear association, cointegration confirms a structural economic link. If the linear combination of these series is stationary, the spread between them is predictable.
This is the bedrock of statistical arbitrage, where traders sell the overvalued asset and buy the undervalued one. When the spread widens beyond historical norms, the cointegration property suggests it will revert.
This relationship is crucial for hedging strategies involving derivatives and underlying tokens. Without cointegration, a spread trade lacks a mean-reverting anchor, making it highly risky.
It allows for the construction of delta-neutral portfolios that rely on price convergence.