Covariance

IntermediateRisk Management2 min read

Quick Definition

A statistical measure indicating the directional relationship between two asset returns — positive when they move together, negative when they move oppositely.

What Is Covariance?

Covariance measures how two assets move in relation to each other. It's the building block for portfolio variance, correlation, and modern portfolio theory.

Understanding Covariance:

CovarianceMeaning
Positive (+)Assets tend to move in the same direction
Negative (-)Assets tend to move in opposite directions
Zero (0)No linear relationship between movements

Covariance vs. Correlation:

FeatureCovarianceCorrelation
ScaleUnbounded (hard to interpret)-1 to +1 (standardized)
UnitsProduct of return unitsUnitless
Ease of UseDifficult to compareEasy to compare
Use in MathPortfolio variance calcQualitative relationship

Relationship: Correlation = Covariance / (σA × σB)

Role in Portfolio Theory: Portfolio Variance = Σ(wi² × σi²) + ΣΣ(wi × wj × Cov(i,j))

The covariance terms often dominate portfolio risk for portfolios with many assets. This is why diversification works — negative or low covariance between holdings reduces total portfolio risk.

Practical Example:

  • Stock A and B both have 20% annual volatility
  • If perfectly correlated (Cov = max): 50/50 portfolio also has 20% volatility
  • If uncorrelated (Cov = 0): 50/50 portfolio has ~14.1% volatility (30% reduction!)
  • If perfectly negatively correlated: 50/50 portfolio has 0% volatility

Formula

Formula

Cov(X,Y) = Σ[(Xi - X̄)(Yi - Ȳ)] / (n-1)

Covariance Example

  • 1Two stocks with positive covariance tend to rise and fall together — adding both doesn't diversify much
  • 2Stocks and bonds historically have negative covariance, which is why 60/40 portfolios reduce risk