Efficient Estimation of Covariance and (Partial) Correlation

This package implements an analytic shrinkage approach for inferring the covariance matrix. The estimator is statistically highly accurate and efficient, applicable to "small n, large p" data, and always returns a positive definite and well-conditioned matrix. Nevertheless, this method requires only little a priori modeling and is computationally cheap. In addition to covariance estimation the package contains similar functions for inferring variances, correlations, partial correlations, partial covariances, and regression coefficients. Furthermore, it provides functions for fast SVD computation, for computing the pseudoinverse, for checking the rank and positive definiteness of a matrix, and for the computationally fast inversion of the covariance and correlation matrix.

Version: 1.4.7

Authors: Juliane Schäfer, Rainer Opgen-Rhein, and Korbinian Strimmer.

Documentation and Installation:

• Help pages and release history.
• Local download of corpcor source package.
• Licensed under the GNU GPL version 3 (or any later version).

Additional Information and Relevant Papers:

• The two-way shrinkage formula for the covariance matrix estimator implemented in "corpcor" (click on image to enlarge):
  .
• An illustrative test script and the two accompanying example data sets smalldata.txt and largedata.txt.
• Correlation Shrinkage: Schäfer, J., and K. Strimmer. 2005. A shrinkage approach to large-scale covariance matrix estimation and implications for functional genomics. Statist. Appl. Genet. Mol. Biol. 4: 32. (reprint)
• Variance Shrinkage: Opgen-Rhein, R., and K. Strimmer. 2007. Accurate ranking of differentially expressed genes by a distribution-free shrinkage approach. Statist. Appl. Genet. Mol. Biol. 6: 9. (reprint)

Back to software page.