nullspace_penalty: Null-space penalty of the precision matrix

View source: R/tmbutils.R

nullspace_penaltyR Documentation

Null-space penalty of the precision matrix

Description

This add e.g., constraints zero intercept and slope for second-order random walk model.

Usage

nullspace_penalty(x)

Arguments

x

a precision matrix

Details

Using eigen decomposition to find the zero eigenvalues, which is then added back to the original penalized matrix. Note that this leads to loss of sparseness.


kklot/ktools documentation built on Aug. 13, 2024, 7:08 p.m.