GFORCE (version 0.1.4)

gforce.Gamma: Estimates \(\Gamma\) for the PECOK SDP.

Description

In particular, it returns in diagonal form the estimator \(\Gamma\) used to construct the PECOK penalized covariance estimator.

Usage

gforce.Gamma(X, par = FALSE, fast_estimator = FALSE, R_only = FALSE)

Arguments

X

\(n x d\) matrix. Each row represents a realization of a \(d\) dimensional random vector.

par

logical expression. If par == TRUE, then a multi-threaded version of the function is called. If par == FALSE, a single-threaded version is called.

fast_estimator

logical expression. If fast_estimator == TRUE, then the alternative estimator for \(\hat \Gamma\) is used.

R_only

logical expression. If R_only == TRUE, then no native code is run. If fast_estimator != TRUE this is ignored.

Value

The estimator \(\Gamma\) as a \(d\) dimensional numeric vector.

References

F. Bunea, C. Giraud, M. Royer and N. Verzelen. PECOK: a convex optimization approach to variable clustering. arXiv:1606.05100, 2016.

Examples

# NOT RUN {
dat <- gforce.generator(5,20,20,3)
gam_hat <- gforce.Gamma(dat$X)

# }