pPPCA_INFO: Penalized profile log-likelihood of a PPPCA model

In WeiAkaneDeng/SPAC2: Statistical and Probabilistic Algorithms for Classification and Clustering

Description

The function returns either the penalized profile log-likelihood or the value that maximizes the penalized profile log-likelihood for a given tuning parameter value.

Usage

pPPCA_INFO(lam, Tvotes = 100, M, AIC = TRUE, tau = 0.001, beta = NULL)

Arguments

lam	a numerical vector of positive sample eigenvalues
Tvotes	the number of possible tuning parameter values to be searched
M	the number of observations used to calculate sample covariance.
AIC	a logical indicator to use AIC as the information criterion, if FALSE, then BIC is used; the default option is AIC.
tau	a tolerance threshold for the smallest eigenvalue, the default value is 0.001.
beta	a numeric between 0 and 1 indicating the weight towards penalty function 1 or 2.

Value

an integer K that maximizes the penalized profile log-likelihood for the given delta value.

WeiAkaneDeng/SPAC2 documentation built on Jan. 15, 2022, 5:01 a.m.