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

Estimation of a precision matrix (i.e. inverse covariance matrix) is widely used to exploit conditional independence among continuous variables. The influence of abnormal observations is exacerbated in a high dimensional setting as the dimensionality increases. In this work, we propose robust estimation of the inverse covariance matrix based on an regularised objective function with a weighted sample covariance matrix. The robustness of the proposed objective function can be justified by a nonparametric technique of the integrated squared error criterion. To address the non-convexity of the objective function, we develop an efficient algorithm in a similar spirit of majorisation-minimisation. Asymptotic consistency of the proposed estimator is also established. The performance of the proposed method is compared with several existing approaches via numerical simulations. We further demonstrate the merits of the proposed method with application in genetic network inference.

[1]  M. C. Jones,et al.  Robust and efficient estimation by minimising a density power divergence , 1998 .

[2]  Charles R. Johnson,et al.  Matrix analysis , 1985, Statistical Inference for Engineers and Data Scientists.

[3]  F. Perron,et al.  Minimax estimators of a covariance matrix , 1992 .

[4]  Po-Ling Loh,et al.  High-dimensional robust precision matrix estimation: Cellwise corruption under $\epsilon$-contamination , 2015, 1509.07229.

[5]  Roland Fried,et al.  On Robust Gaussian Graphical Modeling , 2010 .

[6]  David W. Scott,et al.  Parametric Statistical Modeling by Minimum Integrated Square Error , 2001, Technometrics.

[7]  T. Cai,et al.  A Constrained ℓ1 Minimization Approach to Sparse Precision Matrix Estimation , 2011, 1102.2233.

[8]  C. Croux,et al.  Robust High-Dimensional Precision Matrix Estimation , 2014, 1501.01219.

[9]  H. Zou,et al.  One-step Sparse Estimates in Nonconcave Penalized Likelihood Models. , 2008, Annals of statistics.

[10]  Mathias Drton,et al.  Robust graphical modeling of gene networks using classical and alternative t-distributions , 2010, 1009.3669.

[11]  Eric C. Chi,et al.  Robust Parametric Classification and Variable Selection by a Minimum Distance Criterion , 2011, 1109.6090.

[12]  Jianhua Z. Huang,et al.  Covariance matrix selection and estimation via penalised normal likelihood , 2006 .

[13]  Mark Goadrich,et al.  The relationship between Precision-Recall and ROC curves , 2006, ICML.

[14]  Francisco J. Prieto,et al.  Multivariate Outlier Detection and Robust Covariance Matrix Estimation , 2001, Technometrics.

[15]  Hongzhe Li,et al.  Robust Gaussian Graphical Modeling Via l1 Penalization , 2012, Biometrics.

[16]  D. Hayes.  Pharmacogenomic Predictor of Sensitivity to Preoperative Chemotherapy With Paclitaxel and Fluorouracil, Doxorubicin, and Cyclophosphamide in Breast Cancer , 2007 .

[17]  Christophe Ambroise,et al.  Inferring sparse Gaussian graphical models with latent structure , 2008, 0810.3177.

[18]  Stephen P. Boyd,et al.  Convex Optimization , 2004, Algorithms and Theory of Computation Handbook.

[19]  R. Tibshirani,et al.  Sparse inverse covariance estimation with the graphical lasso. , 2008, Biostatistics.

[20]  D. Hunter,et al.  A Tutorial on MM Algorithms , 2004 .

[21]  M. Yuan,et al.  Model selection and estimation in the Gaussian graphical model , 2007 .

[22]  Michael I. Jordan.  Graphical Models , 2003 .

[23]  Garth Tarr,et al.  Robust estimation of precision matrices under cellwise contamination , 2015, Comput. Stat. Data Anal..

[24]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[25]  Seung-Jean Kim,et al.  Condition‐number‐regularized covariance estimation , 2013, Journal of the Royal Statistical Society. Series B, Statistical methodology.

[26]  Adam J. Rothman,et al.  Sparse estimation of large covariance matrices via a nested Lasso penalty , 2008, 0803.3872.

[27]  René Natowicz,et al.  Prediction of the outcome of preoperative chemotherapy in breast cancer using DNA probes that provide information on both complete and incomplete responses , 2008, BMC Bioinformatics.

[28]  Alexandre d'Aspremont,et al.  Convex optimization techniques for fitting sparse Gaussian graphical models , 2006, ICML.

[29]  Xinwei Deng,et al.  An improved modified cholesky decomposition approach for precision matrix estimation , 2020, Journal of Statistical Computation and Simulation.

[30]  Jessika Weiss,et al.  Graphical Models In Applied Multivariate Statistics , 2016 .

[31]  Xinwei Deng,et al.  Large Gaussian Covariance Matrix Estimation With Markov Structures , 2009 .

[32]  W. Sharpe.  The Sharpe Ratio , 1994 .

[33]  Jianqing Fan,et al.  Sparsistency and Rates of Convergence in Large Covariance Matrix Estimation. , 2007, Annals of statistics.

[34]  Larry A. Wasserman,et al.  High Dimensional Semiparametric Gaussian Copula Graphical Models. , 2012, ICML 2012.

[35]  N. Meinshausen,et al.  High-dimensional graphs and variable selection with the Lasso , 2006, math/0608017.

[36]  M. Pourahmadi.  Maximum likelihood estimation of generalised linear models for multivariate normal covariance matrix , 2000 .

[37]  Olivier Ledoit,et al.  A well-conditioned estimator for large-dimensional covariance matrices , 2004 .

[38]  H. Zou,et al.  Regularization and variable selection via the elastic net , 2005 .

[39]  M. Rudemo.  Empirical Choice of Histograms and Kernel Density Estimators , 1982 .