Correlation measures the degree to which two assets, strategies, or return series move together. It ranges from +1 (perfectly correlated, always moving in the same direction by proportional amounts) to -1 (perfectly inversely correlated, always moving in opposite directions). A correlation of zero means the two series have no linear relationship. Correlation is the foundation of portfolio diversification and is essential for understanding how combining strategies or assets affects overall portfolio risk.
How correlation works
The Pearson correlation coefficient is calculated as the covariance of two return series divided by the product of their individual standard deviations. The result is a dimensionless number between -1 and +1 that indicates both the direction and strength of the linear relationship.
A correlation of +0.8 between two stocks means they tend to move in the same direction with high consistency. When one rises, the other usually rises too. A correlation of -0.5 means they tend to move in opposite directions about half the time. A correlation near zero means knowing one stock's return provides no useful information about the other's return.
Why correlation matters for portfolio construction
The primary benefit of diversification comes from combining assets or strategies with low or negative correlations. When portfolio components are uncorrelated, some positions gain while others lose, smoothing the overall equity curve and reducing drawdowns. The mathematical result is that a portfolio of uncorrelated strategies has lower volatility than the average volatility of its components.
If two strategies each have a Sharpe ratio of 1.0 and are uncorrelated, combining them equally produces a portfolio Sharpe ratio of approximately 1.4 (1.0 multiplied by the square root of 2). This is one of the few "free lunches" in finance: combining uncorrelated return streams improves risk-adjusted returns without requiring better individual strategies.
Correlation instability
A critical challenge is that correlations are not stable over time. Assets that show low correlation during calm markets often become highly correlated during crises, precisely when diversification is most needed. This phenomenon, called correlation breakdown, means that historical correlation estimates may overstate the diversification benefit during market stress.
Rolling correlation analysis, which calculates correlation over a moving window, reveals how the relationship between two assets evolves over time. This is more informative than a single correlation number calculated over the entire history.
Practical example
A trader runs two strategies: a momentum equity strategy and a short-term mean reversion strategy. Over the past three years, their daily return correlation is 0.15, indicating low correlation. The momentum strategy has a Sharpe ratio of 1.2 with 18% maximum drawdown. The mean reversion strategy has a Sharpe ratio of 0.9 with 10% maximum drawdown. An equal-weight combination produces a portfolio Sharpe ratio of 1.5 with a 12% maximum drawdown, better than either strategy alone on both metrics.
How Tektii helps
Tektii calculates correlation matrices between strategies and benchmarks, enabling traders to evaluate diversification benefits before combining strategies into a portfolio. The platform supports rolling correlation analysis to identify periods when correlations change. By providing correlation analytics alongside individual strategy metrics, Tektii helps traders build portfolios that are more resilient than any single strategy.