Adaptive estimation of the copula correlation matrix for semiparametric elliptical copulas

We study the adaptive estimation of copula correlation matrix $\Sigma$ for the semi-parametric elliptical copula model. In this context, the correlations are connected to Kendall's tau through a sine function transformation. Hence, a natural estimate for $\Sigma$ is the plug-in estimator $\hat{\Sigma}$ with Kendall's tau statistic. We first obtain a sharp bound on the operator norm of $\hat{\Sigma}-\Sigma$. Then we study a factor model of $\Sigma$, for which we propose a refined estimator $\widetilde{\Sigma}$ by fitting a low-rank matrix plus a diagonal matrix to $\hat{\Sigma}$ using least squares with a nuclear norm penalty on the low-rank matrix. The bound on the operator norm of $\hat{\Sigma}-\Sigma$ serves to scale the penalty term, and we obtain finite sample oracle inequalities for $\widetilde{\Sigma}$. We also consider an elementary factor copula model of $\Sigma$, for which we propose closed-form estimators. All of our estimation procedures are entirely data-driven.

[1]  E. Schmidt Zur Theorie der linearen und nichtlinearen Integralgleichungen , 1907 .

[2]  C. Eckart,et al.  The approximation of one matrix by another of lower rank , 1936 .

[3]  Herman Rubin,et al.  Statistical Inference in Factor Analysis , 1956 .

[4]  W. Kruskal Ordinal Measures of Association , 1958 .

[5]  W. Hoeffding Probability Inequalities for sums of Bounded Random Variables , 1963 .

[6]  R. Forthofer,et al.  Rank Correlation Methods , 1981 .

[7]  G. Simons,et al.  On the theory of elliptically contoured distributions , 1981 .

[8]  Charles R. Johnson,et al.  Topics in Matrix Analysis , 1991 .

[9]  G. Watson Characterization of the subdifferential of some matrix norms , 1992 .

[10]  M. Kendall,et al.  Rank Correlation Methods , 1949 .

[11]  D. Petz A survey of certain trace inequalities , 1994 .

[12]  A. Sklar,et al.  Random variables, distribution functions, and copulas---a personal look backward and forward , 1996 .

[13]  C. Klaassen,et al.  Efficient estimation in the bivariate normal copula model: normal margins are least favourable , 1997 .

[14]  F. Lindskog,et al.  Multivariate extremes, aggregation and dependence in elliptical distributions , 2002, Advances in Applied Probability.

[15]  S. Kotz,et al.  The Meta-elliptical Distributions with Given Marginals , 2002 .

[16]  A. McNeil,et al.  KENDALL'S TAU FOR ELLIPTICAL DISTRIBUTIONS ∗ , 2003 .

[17]  P. Embrechts,et al.  Chapter 8 – Modelling Dependence with Copulas and Applications to Risk Management , 2003 .

[18]  Defeng Sun,et al.  A Quadratically Convergent Newton Method for Computing the Nearest Correlation Matrix , 2006, SIAM J. Matrix Anal. Appl..

[19]  A. McNeil,et al.  The t Copula and Related Copulas , 2005 .

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

[21]  C. Klüppelberg,et al.  Semi‐Parametric Models for the Multivariate Tail Dependence Function – the Asymptotically Dependent Case , 2008 .

[22]  Larry A. Wasserman,et al.  The Nonparanormal: Semiparametric Estimation of High Dimensional Undirected Graphs , 2009, J. Mach. Learn. Res..

[23]  Claudia Klüppelberg,et al.  Copula structure analysis , 2009 .

[24]  P. Bickel,et al.  Covariance regularization by thresholding , 2009, 0901.3079.

[25]  Harrison H. Zhou,et al.  Optimal rates of convergence for covariance matrix estimation , 2010, 1010.3866.

[26]  V. Koltchinskii,et al.  Nuclear norm penalization and optimal rates for noisy low rank matrix completion , 2010, 1011.6256.

[27]  Pablo A. Parrilo,et al.  Latent variable graphical model selection via convex optimization , 2010, 2010 48th Annual Allerton Conference on Communication, Control, and Computing (Allerton).

[28]  A. Tsybakov,et al.  Estimation of high-dimensional low-rank matrices , 2009, 0912.5338.

[29]  Martin J. Wainwright,et al.  Estimation of (near) low-rank matrices with noise and high-dimensional scaling , 2009, ICML.

[30]  Xi Luo High Dimensional Low Rank and Sparse Covariance Matrix Estimation via Convex Minimization , 2011 .

[31]  Martin J. Wainwright,et al.  Noisy matrix decomposition via convex relaxation: Optimal rates in high dimensions , 2011, ICML.

[32]  M. Wegkamp,et al.  Optimal selection of reduced rank estimators of high-dimensional matrices , 2010, 1004.2995.

[33]  Pablo A. Parrilo,et al.  Rank-Sparsity Incoherence for Matrix Decomposition , 2009, SIAM J. Optim..

[34]  Sham M. Kakade,et al.  Robust Matrix Decomposition With Sparse Corruptions , 2011, IEEE Transactions on Information Theory.

[35]  Emmanuel J. Candès,et al.  Exact Matrix Completion via Convex Optimization , 2008, Found. Comput. Math..

[36]  A. Tsybakov,et al.  Comment: "Minimax estimation of large covariance matrices under ℓ1-norm'' , 2012 .

[37]  H. Zou,et al.  Positive-Definite ℓ1-Penalized Estimation of Large Covariance Matrices , 2012, 1208.5702.

[38]  Joel A. Tropp,et al.  User-Friendly Tail Bounds for Sums of Random Matrices , 2010, Found. Comput. Math..

[39]  F. Bunea,et al.  On the sample covariance matrix estimator of reduced effective rank population matrices, with applications to fPCA , 2012, 1212.5321.

[40]  Harrison H. Zhou,et al.  MINIMAX ESTIMATION OF LARGE COVARIANCE MATRICES UNDER ℓ1-NORM , 2012 .

[41]  Roman Vershynin,et al.  Introduction to the non-asymptotic analysis of random matrices , 2010, Compressed Sensing.

[42]  Tong Zhang,et al.  A General Framework of Dual Certificate Analysis for Structured Sparse Recovery Problems , 2012, 1201.3302.

[43]  Harrison H. Zhou,et al.  OPTIMAL RATES OF CONVERGENCE FOR SPARSE COVARIANCE MATRIX ESTIMATION , 2012, 1302.3030.

[44]  H. Zou,et al.  Regularized rank-based estimation of high-dimensional nonparanormal graphical models , 2012, 1302.3082.

[45]  Pablo A. Parrilo,et al.  Diagonal and Low-Rank Matrix Decompositions, Correlation Matrices, and Ellipsoid Fitting , 2012, SIAM J. Matrix Anal. Appl..

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

[47]  J. Tropp User-Friendly Tools for Random Matrices: An Introduction , 2012 .

[48]  Karim Lounici High-dimensional covariance matrix estimation with missing observations , 2012, 1201.2577.

[49]  Fang Han,et al.  Transelliptical Graphical Models , 2012, NIPS.

[50]  Xi Luo,et al.  N ov 2 01 1 High Dimensional Low Rank and Sparse Covariance Matrix Estimation via Convex Minimization ∗ , 2011 .

[51]  Han Liu,et al.  Optimal Rates of Convergence for Latent Generalized Correlation Matrix Estimation in Transelliptical Distribution , 2013 .

[52]  Jianqing Fan,et al.  Large covariance estimation by thresholding principal orthogonal complements , 2011, Journal of the Royal Statistical Society. Series B, Statistical methodology.

[53]  Han Liu,et al.  Optimal Rates of Convergence of Transelliptical Component Analysis , 2013 .

[54]  Cun-Hui Zhang,et al.  Multivariate Analysis of Nonparametric Estimates of Large Correlation Matrices , 2014, 1403.6195.

[55]  Joel A. Tropp,et al.  An Introduction to Matrix Concentration Inequalities , 2015, Found. Trends Mach. Learn..

[56]  Han Liu,et al.  Statistical analysis of latent generalized correlation matrix estimation in transelliptical distribution. , 2013, Bernoulli : official journal of the Bernoulli Society for Mathematical Statistics and Probability.