prox_gradient_mapping: Proximal-gradient mapping method.

In changepointsHD: Change-Point Estimation for Expensive and High-Dimensional Models

Description

Performs the proximal-gradient mapping operation to estimate a regularized version of the inverse cov. matrix. Follows the procedure described in, http://dept.stat.lsa.umich.edu/~yvesa/sto_prox.pdf

Usage

prox_gradient_mapping(data, theta_start, update_w, update_change, regularizer,
  max_iter, tol)

Arguments

data	N x P matrix corresponding to the raw data.
theta_start	Initial value for precision estimate.
update_w	Step size for prox-gradient mapping.
update_change	Proportion of update_w to keep when the algorithm fails to successfully estimate precision.
regularizer	Regularizing constant, lambda.
max_iter	Number of mapping iterations.
tol	Tolerance at which the algorithm stops running.

Value

Theta (precision matrix) estimate.

Author(s)

Leland Bybee <[e-mail]>

changepointsHD documentation built on Sept. 20, 2019, 5:04 p.m.