Shrinkage Estimation

Below  is the report I have written with my classmate Vinod B and Christian H on covariance shrinkage estimation and application to portfolio optimization.

Overview of Shrinkage Estimation and its Benefits
The covariance matrix of returns Σ plays a central role in Markowitz portfolio theory; however, in practice it is unknown and must be estimated from historical data. Although the sample covariance matrix Σ-hat is an unbiased estimator of Σ, it is extremely unstable when the number of parameters is large compared to the number of returns. One approach to portfolio optimization replaces the unknown Σ with a shrinkage estimator. Ledoit and Wolf (2003) propose to shrink towards structured matrices to produce an estimator with relatively small error in comparison to Σ-hat.

In the report, we explored the advantage of using shrinkage estimation, its application, methodology, and results. Please see the full report Shrinkage Estimation- Vinod B, Chris H, Sharon L.

%d bloggers like this: