| 1 |
Package: corpcor |
| 2 |
Version: 1.6.10 |
| 3 |
Date: 2021-09-16 |
| 4 |
Title: Efficient Estimation of Covariance and (Partial) Correlation |
| 5 |
Author: Juliane Schafer, Rainer Opgen-Rhein, Verena Zuber, Miika Ahdesmaki, |
| 6 |
A. Pedro Duarte Silva, and Korbinian Strimmer. |
| 7 |
Maintainer: Korbinian Strimmer <[email protected]> |
| 8 |
Depends: R (>= 3.0.2) |
| 9 |
Imports: stats |
| 10 |
Suggests: |
| 11 |
Description: Implements a James-Stein-type shrinkage estimator for |
| 12 |
the covariance matrix, with separate shrinkage for variances and correlations. |
| 13 |
The details of the method are explained in Schafer and Strimmer (2005) |
| 14 |
<DOI:10.2202/1544-6115.1175> and Opgen-Rhein and Strimmer (2007) |
| 15 |
<DOI:10.2202/1544-6115.1252>. The approach is both computationally as well |
| 16 |
as statistically very efficient, it is applicable to "small n, large p" data, |
| 17 |
and always returns a positive definite and well-conditioned covariance matrix. |
| 18 |
In addition to inferring the covariance matrix the package also provides |
| 19 |
shrinkage estimators for partial correlations and partial variances. |
| 20 |
The inverse of the covariance and correlation matrix |
| 21 |
can be efficiently computed, as well as any arbitrary power of the |
| 22 |
shrinkage correlation matrix. Furthermore, functions are available for fast |
| 23 |
singular value decomposition, for computing the pseudoinverse, and for |
| 24 |
checking the rank and positive definiteness of a matrix. |
| 25 |
License: GPL (>= 3) |
| 26 |
URL: https://strimmerlab.github.io/software/corpcor/ |
| 27 |
NeedsCompilation: no |
| 28 |
Packaged: 2021-09-16 14:05:52 UTC; strimmer |
| 29 |
Repository: CRAN |
| 30 |
Date/Publication: 2021-09-16 15:30:08 UTC |