• Corpus ID: 245853871

Robust graphical lasso based on multivariate Winsorization

@inproceedings{Lafit2022RobustGL,
  title={Robust graphical lasso based on multivariate Winsorization},
  author={Ginette Lafit and Francisco J. Nogales and Marcelo Ruiz and Ruben H. Zamar},
  year={2022}
}
We propose the use of a robust covariance estimator based on multivariate Winsorization in the context of the Tarr–Müller–Weber framework for sparse estimation of the precision matrix of a Gaussian graphical model. Likewise Croux–Öllerer’s precision matrix estimator, our proposed estimator attains the maximum finite sample breakdown point of 0.5 under cellwise contamination. We conduct an extensive Monte Carlo simulation study to assess the performance of ours and the currently existing… 

The Influence Function of Graphical Lasso Estimators

This paper studiesarse estimation procedures for precision matrices such as the graphical lasso theoretically, by deriving and com-paring their influence function, sensitivity curves and asymptotic variances.

References

SHOWING 1-10 OF 33 REFERENCES

Robust estimation of sparse precision matrix using adaptive weighted graphical lasso approach

The robustness of the proposed objective function can be justified by a nonparametric technique of the integrated squared error criterion, and an efficient algorithm is developed in a similar spirit of majorisation-minimisation to address the non-convexity of the objective function.

NETWORK EXPLORATION VIA THE ADAPTIVE LASSO AND SCAD PENALTIES.

Non-concave penalties and the adaptive LASSO penalty are introduced to attenuate the bias problem in the network estimation to solve the problem of precision matrix estimation.

Sparse inverse covariance estimation with the graphical lasso.

Using a coordinate descent procedure for the lasso, a simple algorithm is developed that solves a 1000-node problem in at most a minute and is 30-4000 times faster than competing methods.

Robust estimation of precision matrices under cellwise contamination

Handling Cellwise Outliers by Sparse Regression and Robust Covariance

A detection-imputation method is constructed which alternates between flagging outlying cells and updating the covariance matrix as in the EM algorithm for detecting cellwise outliers.

Inferring sparse Gaussian graphical models with latent structure

This work describes a novel framework taking into account a latent structure on the concentration matrix that is used to drive a penalty matrix and thus to recover a graphical model with a constrained topology.

Robust estimation of multivariate location and scatter in the presence of cellwise and casewise contamination

The need for a new generation of robust estimators that can efficiently deal with cellwise outliers and at the same time show good performance under casewise outliers is highlighted.

A robust partial correlation measure

A robust, efficient measure of partial association which uses residuals based on rank estimates of regression parameters is proposed. The estimates of regression coefficients are based on the

Propagation of outliers in multivariate data

It is shown that standard high-breakdown affine equivariant estimators propagate outliers and therefore show poor breakdown behavior under componentwise contamination when the dimension d is high.

ROBUST ESTIMATES, RESIDUALS, AND OUTLIER DETECTION WITH MULTIRESPONSE DATA

An overview of concepts and techniques pertaining to (i) the robust estimation of multivariate location and dispersion; (ii) the analysis of two types of multidimensional residuals; and (iii) the detection of multiresponse outliers.