Statistics

Covariance Matrix

The matrix of pairwise covariances between a set of variables, central to portfolio risk.

Symmetric and positive semi-definite. Portfolio variance is w'Σw, so the whole of modern portfolio construction runs through it.

Positive semi-definiteness is a real constraint, not a technicality: it is why three assets cannot all be pairwise correlated at -0.9. An interviewer asking whether a given matrix is a valid correlation matrix is testing exactly this.

The estimation problem. With n assets and T observations, the sample estimate is badly conditioned when T is not much larger than n, and is singular when T < n. Shrinkage toward a structured target is the standard fix.

Related terms

Practise this

Put it into practice

Knowing the definition is not the same as spotting where it applies under time pressure. Work the question bank free.

Start practising free

Browse the full quant interview glossary