AI & Quantitative

Correlation

A statistical measure ranging from -1 to +1 that describes how closely two assets' price movements are related to each other.

Explained Simply

Correlation quantifies the relationship between two assets' returns. A correlation of +1.0 means they move in perfect lockstep; -1.0 means they move in exactly opposite directions; 0 means no relationship. Most stocks have positive correlations (0.3 to 0.7) with each other because they're influenced by the same macroeconomic factors. During market crashes, correlations spike toward 1.0 as panic selling hits everything — this is called "correlation convergence" and it undermines diversification precisely when you need it most. Understanding correlation is essential for portfolio construction: holding 10 stocks with 0.9 correlation gives much less diversification than holding 5 stocks with 0.3 correlation. Pairs traders specifically seek highly correlated assets that temporarily diverge.

How Tradewink Uses Correlation

Tradewink's PairsTrader uses correlation analysis to identify and monitor pairs trading candidates — stocks with historically high correlation whose prices have temporarily diverged. The PortfolioRiskAnalyzer tracks cross-asset correlations within the portfolio to ensure true diversification and warns when portfolio correlation is dangerously concentrated. The CointegrationAnalyzer goes beyond simple correlation to test for cointegration — a stronger statistical relationship that accounts for non-stationary price data.

Related Terms

Learn More

See Correlation in action

Tradewink uses correlation as part of its AI trading signal pipeline. Start getting signals that use this concept to find real opportunities.