Correlation
Correlation measures how closely two assets' returns move together, ranging from +1 (perfectly in sync) through 0 (unrelated) to -1 (perfectly opposite).
Correlation is a normalized statistic, usually the Pearson correlation coefficient, that captures the strength and direction of the linear relationship between two return series. It is calculated as the covariance of the two assets divided by the product of their individual standard deviations, which rescales the result to lie between -1 and +1. A value of +1 means the two assets move in lockstep, -1 means they move exactly opposite, and 0 means there is no linear relationship.
Correlation is the mathematical backbone of diversification. Combining assets that are not perfectly correlated reduces portfolio volatility without necessarily sacrificing expected return, because losses in one holding are partly offset by others. The lower (or more negative) the correlations among holdings, the greater the diversification benefit, which is why correlation matrices sit at the heart of portfolio construction and risk models.
Two cautions matter in practice. First, correlation is not causation and only captures linear co-movement, so it can miss nonlinear or tail dependencies. Second, correlations are not stable: assets that appear diversified in calm markets often spike toward +1 during crises, precisely when diversification is needed most. Robust portfolio methods account for this instability rather than assuming historical correlations will persist.
hedgewing.ai relies on correlation in its allocation engine through hierarchical risk parity, which uses the correlation structure among assets to build diversified, risk-balanced portfolios that are more robust to estimation error than naive mean-variance optimization. Correlation-based features also inform the ensemble's view of how individual names relate to broader market and factor movements.
Related terms
Standard Deviation · Diversification · Hierarchical Risk Parity (HRP) · Beta
Back to the hedgewing.ai glossary · See AI stock forecasts