NLL: Evaluate the (penalized) (fused) likelihood

In CFWP/rags2ridges: Ridge Estimation of Precision Matrices from High-Dimensional Data

Evaluate the (penalized) (fused) likelihood

Description

Functions that evaluate the (penalized) (fused) likelihood.

Usage

NLL(S, P)

PNLL(S, P, T, lambda)

NLL.fused(Slist, Plist, ns)

PNLL.fused(Slist, Plist, ns, Tlist, lambda)

Arguments

S, Slist	A (list of) positive semi definite sample covariance matrices.
P, Plist	A (list of) positive definite precision matrices.
T, Tlist	A (list of) positive definite target matrices.
lambda	A numeric penalty parameter. For the .fused functions, this is a penalty matrix.
ns	A numeric of sample sizes.

Value

A single number.

Author(s)

Anders Ellern Bilgrau, Carel F.W. Peeters <[e-mail]>, Wessel N. van Wieringen

See Also

ridgeP, ridgeP.fused

Examples

ns <- c(4,5)
Slist <- createS(n = ns, p = 5)
Plist <- list(diag(5), diag(2,5))
Tlist <- list(diag(5), diag(5))

NLL(Slist[[1]], Plist[[1]])
PNLL(Slist[[1]], Plist[[1]], Tlist[[1]], lambda = 1)
NLL.fused(Slist, Plist, ns)
PNLL.fused(Slist, Plist, ns, Tlist, lambda = diag(2))

CFWP/rags2ridges documentation built on Oct. 21, 2023, 10:19 a.m.