WeiAkaneDeng/SPAC2: Statistical and Probabilistic Algorithm for Classification and Clustering

This package provides a statistical algorithm that implements a penalized profile log-likelihood criterion to estimate the number of effective dimensions of a data matrix. The data structure is modeled similarly as in Probabilistic Principal Components Analysis (PPCA). The package also provides various functions to simulate either the sample eigenvalue or the data matrix under specific covariance structures and possibly with violation to normality or independence assumption. The effective dimension or the number of principal components uncovered using our approach can be then used as inputs in subsequent analysis such as clustering or classification.

Getting started

Package details

AuthorWei Q. Deng, Radu R. Craiu
MaintainerWei Q. Deng <[email protected]>
LicenseGPL-2
Version0.9.3
Package repositoryView on GitHub
Installation Install the latest version of this package by entering the following in R:
install.packages("devtools")
library(devtools)
install_github("WeiAkaneDeng/SPAC2")
WeiAkaneDeng/SPAC2 documentation built on April 7, 2018, 10:07 a.m.