zrmacc/MGMM: Missingness Aware Gaussian Mixture Models

Parameter estimation for Gaussian Mixture Models (GMMs) allowing for missingness in the input vectors. Rather than imputing the missing inputs and clustering the completed data, this package uses the EM algorithm to accommodate both the missing inputs and the missing cluster assignments. Provides maximum likelihood estimates of model parameters, maximum a posteriori classifications of the inputs, and a completed version of the input data, with missing values replaced by their posterior expectations.

Getting started

Package details

Package repositoryView on GitHub
Installation Install the latest version of this package by entering the following in R:
zrmacc/MGMM documentation built on May 13, 2019, 10:31 a.m.