Sparse estimation of high-dimensional correlation matrices

Ying Cui, Chenlei Leng, Defeng Sun

Research output: Contribution to journalArticlepeer-review

27 Scopus citations

Abstract

Several attempts to estimate covariance matrices with sparsity constraints have been made. A convex optimization formulation for estimating correlation matrices as opposed to covariance matrices is proposed. An efficient accelerated proximal gradient algorithm is developed, and it is shown that this method gives a faster rate of convergence. An adaptive version of this approach is also discussed. Simulation results and an analysis of a cardiovascular microarray confirm its performance and usefulness.

Original languageEnglish (US)
Pages (from-to)390-403
Number of pages14
JournalComputational Statistics and Data Analysis
Volume93
DOIs
StatePublished - Jan 1 2016
Externally publishedYes

Bibliographical note

Funding Information:
The research of Sun was supported in part by the NUS Academic Research Fund under grant R-146-000-149-112 . We are grateful to two reviewers for their constructive comments.

Publisher Copyright:
© 2014 Elsevier B.V. All rights reserved.

Keywords

  • Accelerated proximal gradient
  • Correlation matrix
  • High-dimensionality
  • Positive definiteness
  • Sparsity

Fingerprint

Dive into the research topics of 'Sparse estimation of high-dimensional correlation matrices'. Together they form a unique fingerprint.

Cite this