Regularized rank-based estimation of high-dimensional nonparanormal graphical models

DOI10.1214/12-AOS1041zbMATH Open1373.62138arXiv1302.3082MaRDI QIDQ741796FDOQ741796

Authors: Lingzhou Xue, Hui Zou 

Publication date: 15 September 2014

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: A sparse precision matrix can be directly translated into a sparse Gaussian graphical model under the assumption that the data follow a joint normal distribution. This neat property makes high-dimensional precision matrix estimation very appealing in many applications. However, in practice we often face nonnormal data, and variable transformation is often used to achieve normality. In this paper we consider the nonparanormal model that assumes that the variables follow a joint normal distribution after a set of unknown monotone transformations. The nonparanormal model is much more flexible than the normal model while retaining the good interpretability of the latter in that each zero entry in the sparse precision matrix of the nonparanormal model corresponds to a pair of conditionally independent variables. In this paper we show that the nonparanormal graphical model can be efficiently estimated by using a rank-based estimation scheme which does not require estimating these unknown transformation functions. In particular, we study the rank-based graphical lasso, the rank-based neighborhood Dantzig selector and the rank-based CLIME. We establish their theoretical properties in the setting where the dimension is nearly exponentially large relative to the sample size. It is shown that the proposed rank-based estimators work as well as their oracle counterparts defined with the oracle data. Furthermore, the theory motivates us to consider the adaptive version of the rank-based neighborhood Dantzig selector and the rank-based CLIME that are shown to enjoy graphical model selection consistency without assuming the irrepresentable condition for the oracle and rank-based graphical lasso. Simulated and real data are used to demonstrate the finite performance of the rank-based estimators.

Full work available at URL: https://arxiv.org/abs/1302.3082

Recommendations

• High-dimensional rank-based graphical models for non-Gaussian functional data
• Bayesian analysis of nonparanormal graphical models using rank-likelihood
• Estimation of high-dimensional graphical models using regularized score matching
• RANK: Large-Scale Inference With Graphical Nonlinear Knockoffs
• Estimation of nonparanormal graphical models based on ranked set sampling (RSS)
• The nonparanormal: semiparametric estimation of high dimensional undirected graphs
• Nonparametric and high-dimensional functional graphical models
• High-dimensional rank-based inference
• Direct estimation of differential networks under high-dimensional nonparanormal graphical models
• Inference for Nonparanormal Partial Correlation via Regularized Rank-Based Nodewise Regression

zbMATH Keywords

rate of convergencegraphical LassoDantzig selectorvariable transformationCLIMEnonparanormal graphical model

Mathematics Subject Classification ID

Asymptotic properties of parametric estimators (62F12) Nonparametric estimation (62G05) Asymptotic properties of nonparametric inference (62G20) Ridge regression; shrinkage estimators (Lasso) (62J07)

Cites Work

• Introduction to Graphical Modelling
• The Adaptive Lasso and Its Oracle Properties
• Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties
• Title not available (Why is that?)
• Title not available (Why is that?)
• High-dimensional semiparametric Gaussian copula graphical models
• Simultaneous analysis of Lasso and Dantzig selector
• The Dantzig selector: statistical estimation when \(p\) is much larger than \(n\). (With discussions and rejoinder).
• High-dimensional graphs and variable selection with the Lasso
• The nonparanormal: semiparametric estimation of high dimensional undirected graphs
• Multiple testing and error control in Gaussian graphical model selection
• Multivariate Dispersion Models Generated From Gaussian Copula
• Title not available (Why is that?)
• Regularized rank-based estimation of high-dimensional nonparanormal graphical models
• Sparsistency and rates of convergence in large covariance matrix estimation
• Sparse inverse covariance estimation with the graphical lasso
• Sparse permutation invariant covariance estimation
• Estimation of copula-based semiparametric time series models
• Model selection through sparse maximum likelihood estimation for multivariate Gaussian or binary data
• Gradient directed regularization for sparse Gaussian concentration graphs, with applications to inference of genetic networks
• Model selection and estimation in the Gaussian graphical model
• Title not available (Why is that?)
• Title not available (Why is that?)
• Partial correlation estimation by joint sparse regression models
• A Class of Statistics with Asymptotically Normal Distribution
• High-dimensional covariance estimation by minimizing \(\ell _{1}\)-penalized log-determinant divergence
• High dimensional inverse covariance matrix estimation via linear programming
• A constrained \(\ell _{1}\) minimization approach to sparse precision matrix estimation
• DASSO: Connections Between the Dantzig Selector and Lasso
• The adaptive and the thresholded Lasso for potentially misspecified models (and a lower bound for the Lasso)
• Ordinal Measures of Association
• Model selection for Gaussian concentration graphs
• Title not available (Why is that?)
• High-dimensional covariance estimation based on Gaussian graphical models
• Efficient Estimation of Semiparametric Multivariate Copula Models
• Robust estimation and outlier detection with correlation coefficients

Cited In (98)

• Mediation analysis for survival data using semiparametric probit models
• High-dimensional semiparametric Gaussian copula graphical models
• Direct shrinkage estimation of large dimensional precision matrix
• High dimensional discrimination analysis via a semiparametric model
• An improved modified cholesky decomposition approach for precision matrix estimation
• Kernel-based tests for joint independence
• Robust sparse Gaussian graphical modeling
• On principal graphical models with application to gene network
• Sparse semiparametric discriminant analysis
• High-dimensional robust precision matrix estimation: cellwise corruption under \(\epsilon \)-contamination
• The cluster graphical Lasso for improved estimation of Gaussian graphical models
• Multiple testing of submatrices of a precision matrix with applications to identification of between pathway interactions
• On integer and MPCC representability of affine sparsity
• Simultaneous inference for pairwise graphical models with generalized score matching
• Detection of block-exchangeable structure in large-scale correlation matrices
• Inference for semiparametric Gaussian copula model adjusted for linear regression using residual ranks
• The nonparanormal: semiparametric estimation of high dimensional undirected graphs
• Hypothesis test of mediation effect in causal mediation model with high-dimensional continuous mediators
• Learning causal networks via additive faithfulness
• Covariance-based sample selection for heterogeneous data: applications to gene expression and autism risk gene detection
• Structure estimation for discrete graphical models: generalized covariance matrices and their inverses
• Estimation and inference in a high-dimensional semiparametric Gaussian copula vector autoregressive model
• Rank-based tapering estimation of bandable correlation matrices
• Diagonal nonlinear transformations preserve structure in covariance and precision matrices
• PC algorithm for nonparanormal graphical models
• On an additive semigraphoid model for statistical networks with application to pathway analysis
• Title not available (Why is that?)
• Estimating finite mixtures of ordinal graphical models
• A sequential scaled pairwise selection approach to edge detection in nonparanormal graphical models
• Limiting spectral distribution of large dimensional Spearman's rank correlation matrices
• On skewed Gaussian graphical models
• High-dimensional rank-based graphical models for non-Gaussian functional data
• Estimation of nonparanormal graphical models based on ranked set sampling (RSS)
• Inference for Nonparanormal Partial Correlation via Regularized Rank-Based Nodewise Regression
• Using Bayesian latent Gaussian graphical models to infer symptom associations in verbal autopsies
• Adjacency-based regularization for partially ranked data with non-ignorable missing
• An Expectation Conditional Maximization Approach for Gaussian Graphical Models
• Rejoinder: ``Robust Bayesian graphical modeling using Dirichlet \(t\)-distributions
• Confidence regions for entries of a large precision matrix
• Robust and sparse estimation of the inverse covariance matrix using rank correlation measures
• Gaussian graphical model estimation with false discovery rate control
• High-dimensional semiparametric bigraphical models
• High-dimensional Bayesian inference in nonparametric additive models
• Sufficient forecasting using factor models
• Scale-invariant sparse PCA on high-dimensional meta-elliptical data
• High dimensional Gaussian copula graphical model with FDR control
• Direct estimation of differential networks under high-dimensional nonparanormal graphical models
• Copula Gaussian Graphical Models for Functional Data
• Post-regularization inference for time-varying nonparanormal graphical models
• A semiparametric graphical modelling approach for large-scale equity selection
• A fast iterative algorithm for high-dimensional differential network
• Nonparametric and high-dimensional functional graphical models
• Regularized rank-based estimation of high-dimensional nonparanormal graphical models
• A review of Gaussian Markov models for conditional independence
• Adaptive estimation of the copula correlation matrix for semiparametric elliptical copulas
• On generating random Gaussian graphical models
• Estimation of high-dimensional graphical models using regularized score matching
• A convex optimization approach to high-dimensional sparse quadratic discriminant analysis
• Estimating sparse precision matrix: optimal rates of convergence and adaptive estimation
• Uniform joint screening for ultra-high dimensional graphical models
• Alternating Direction Methods for Latent Variable Gaussian Graphical Model Selection
• Gaussian graphical model‐based heterogeneity analysis via penalized fusion
• Bayesian analysis of nonparanormal graphical models using rank-likelihood
• Rejoinder of ``Estimating structured high-dimensional covariance and precision matrices: optimal rates and adaptive estimation
• ROCKET: robust confidence intervals via Kendall's tau for transelliptical graphical models
• Conditional score matching for high-dimensional partial graphical models
• Testing the differential network between two gaussian graphical models with false discovery rate control
• Controlling the false discovery rate by a latent Gaussian copula knockoff procedure
• Power enhancement for testing multi-factor asset pricing models via Fisher's method
• An Additive Graphical Model for Discrete Data
• High-dimensional covariance matrix estimation
• Coordinatewise Gaussianization: Theories and Applications
• Nonparametric Functional Graphical Modeling Through Functional Additive Regression Operator
• High-dimensional undirected graphical models for arbitrary mixed data
• Inference on Multi-level Partial Correlations Based on Multi-subject Time Series Data
• High-dimensional inference for cluster-based graphical models
• A unified precision matrix estimation framework via sparse column-wise inverse operator under weak sparsity
• Multiple and multilevel graphical models
• Localizing differentially evolving covariance structures via scan statistics
• Estimation of graphical models: an overview of selected topics
• Discussion of “A Tuning-Free Robust and Efficient Approach to High-Dimensional Regression”
• Integrating approximate single factor graphical models
• Robust Covariance Matrix Estimation for High-Dimensional Compositional Data with Application to Sales Data Analysis
• Robust Shape Matrix Estimation for High-Dimensional Compositional Data with Application to Microbial Inter-Taxa Analysis
• Nonparametric Finite Mixture of Gaussian Graphical Models
• Transfer learning in high-dimensional semiparametric graphical models with application to brain connectivity analysis
• Causal discoveries for high dimensional mixed data
• A novel robust estimation for high-dimensional precision matrices
• On semiparametric exponential family graphical models
• Automatic odor prediction for electronic nose
• Kernel partial correlation: a novel approach to capturing conditional independence in graphical models for noisy data
• Global–local shrinkage multivariate logit-beta priors for multiple response-type data
• Bayesian robust learning in chain graph models for integrative pharmacogenomics
• Robust covariance estimation for high-dimensional compositional data with application to microbial communities analysis
• Joint Bayesian analysis of multiple response-types using the hierarchical generalized transformation model
• Nonparanormal graph quilting with applications to calcium imaging
• Discussion of ``Estimating structured high-dimensional covariance and precision matrices: optimal rates and adaptive estimation
• Model-Based Clustering of Nonparametric Weighted Networks With Application to Water Pollution Analysis

Uses Software

• glasso
• MIM

This page was built for publication: Regularized rank-based estimation of high-dimensional nonparanormal graphical models

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q741796)