High-Dimensional Covariance Estimation when Tails are Heavy

R-NL: Robust Nonlinear Shrinkage

In this article, I discuss a new covariance estimation method from our recent paper “R-NL: Fast and Robust Covariance Estimation for Elliptical Distributions in High Dimensions’’ on arXiv. I introduce the problem we are solving, try to give some intuition on how we are solving it, and briefly present the simple code we developed. On the way, I touch upon some interesting concepts, like the (robust) “Tyler’s estimator”, which I feel are somewhat underused in …

covariance covariance-matrix data science high-dimensional-data statistics thoughts-and-theory