Quick definition Correlation measures the degree to which two variables move together. Positive correlation means they move in the same direction. Negative correlation means they move in opposite directions. What it is Correlation ranges from -1 to +1. A correlation of +0.8 between two stocks means they tend to move together (when one is up, the other is usually up). A correlation of -0.6 means inverse relationship (when one is up, the other tends to be down). A correlation of 0 means no relationship. Correlation is calculated from historical returns. Traders measure correlation over different time periods: 1-month correlation, 1-year correlation, etc. Why it matters Correlation is fundamental to diversification. If two assets are uncorrelated, holding both reduces portfolio risk (diversification benefit). If two assets are highly correlated, holding both provides less diversification. Correlation also drives pairs trading. If two correlated assets diverge, traders bet on convergence, as the pair should revert to historical correlation. Correlation versus Covariance Covariance is the raw measure of joint movement. Correlation is the standardized version (removes units). Two highly correlated assets have high covariance; two uncorrelated assets have covariance near zero. Practical example Stock A and Stock B historically have correlation 0.7. An investor holds equal amounts of both. The portfolio is less diversified than holding A and an uncorrelated asset, but more diversified than holding just A. Now correlations break down in a crash. Both A and B fall 20 percent (correlation becomes closer to 1). The diversification benefit disappears. This is why "correlation goes to 1 in a crisis"—risk-off periods cause all risky assets to fall together. Rolling Correlation Correlation changes over time. Recent correlation might differ from historical correlation. Traders compute rolling correlation (say, 30-day correlation updated daily) to track whether correlations are changing. Microstructure Implications In pairs trading and statistical arbitrage, correlation breakdown is a trading opportunity. If two historically correlated assets diverge, traders exploit the convergence. See also - Covariance - Variance - Pairs Trading - Portfolio Diversification
