"Comparison of Linear Shrinkage Estimators of a Large Covariance Matrix in Normal and Non-normal Distributions"

The problem of estimating the large covariance matrix of both normal and non-normal distributions is addressed. In convex combinations of the sample covariance　matrix and the identity matrix multiplied by a scalor statistic, we suggest a new　estimator of the optimal weight based on exact or approximately unbiased estimators of the numerator and denominator of the optimal weight in non-normal cases.　　It is also demonstrated that the estimators given in the literature have second-order biases. It is numerically shown that the proposed estimator has a good risk　performance. --