Back to the glossary

    methodology

    Ledoit-Wolf Shrinkage

    A method that smooths a noisy covariance matrix toward a structured target, producing more reliable optimizer inputs.

    Sample covariance matrices are noisy when the number of assets is large relative to history. Ledoit-Wolf shrinks the sample matrix toward a constant-correlation or identity target with a data-driven blend coefficient.

    Related terms

    Nested Clustered Optimization (NCO)Correlation

    Get full access to GNG Research

    Create a free account to access portfolio tracking, advanced tools, and more.

    Create Free AccountLog In