Keyboard shortcuts

Press ← or → to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Covariance estimation

solow-covariance covers empirical, shrinkage (Ledoit-Wolf, OAS, arbitrary), robust minimum covariance determinant, sparse inverse via graphical lasso, and elliptic-envelope outlier detection.

Estimators

EstimatorWhat it does
EmpiricalCovarianceSample covariance with an optional store-precision flag. Log-likelihood, Mahalanobis distance, error norm.
ShrunkCovarianceConvex shrinkage toward a scaled identity target with a caller-set intensity.
LedoitWolfLedoit-Wolf (2004) optimal shrinkage intensity estimator.
OasOracle Approximating Shrinkage (Chen 2010), better for small samples.
MinCovDetRousseeuw FAST-MCD robust covariance with consistency correction.
GraphicalLassoL1-penalized inverse covariance for sparse precision (partial correlations).
EllipticEnvelopeRobust ellipsoidal outlier detector built on top of MinCovDet.

Example

#![allow(unused)]
fn main() {
use solow_covariance::LedoitWolf;

let lw = LedoitWolf::new().fit(&x)?;
let cov = &lw.covariance;
let shrinkage = lw.shrinkage;
Ok::<_, solow_core::Error>(())
}

References

  • Ledoit, O., & Wolf, M. (2004). A well-conditioned estimator for large-dimensional covariance matrices. Journal of Multivariate Analysis.
  • Chen, Y., Wiesel, A., Eldar, Y. C., & Hero, A. O. (2010). Shrinkage algorithms for MMSE covariance estimation. IEEE TSP.
  • Rousseeuw, P. J., & Van Driessen, K. (1999). A fast algorithm for the minimum covariance determinant estimator. Technometrics.
  • Friedman, J., Hastie, T., & Tibshirani, R. (2008). Sparse inverse covariance estimation with the graphical lasso. Biostatistics.