FitMix

Given a matrix of random vectors, estimates the parameters for a mixture of
multivariate normal distributions. Accommodates arbitrary patterns of
missingness, provided the elements are missing at random (MAR).

Parameter estimation and classification for Gaussian Mixture Models (GMMs) in the presence of missing data. This package complements existing implementations by allowing for both missing elements in the input vectors and full (as opposed to strictly diagonal) covariance matrices. Estimation is performed using an expectation conditional maximization algorithm that accounts for missingness of both the cluster assignments and the vector components. The output includes the marginal cluster membership probabilities; the mean and covariance of each cluster; the posterior probabilities of cluster membership; and a completed version of the input data, with missing values imputed to their posterior expectations. For additional details, please see McCaw ZR, Julienne H, Aschard H. "Fitting Gaussian mixture models on incomplete data." <doi:10.1186/s12859-022-04740-9>.

Zachary McCaw

MGMM

Missingness Aware Gaussian Mixture Models

FitMix function

<dl><dt>data</dt>
<dd>Numeric data matrix.</dd>
<dt>k</dt>
<dd>Number of mixture components. Defaults to 2.</dd>
<dt>init_means</dt>
<dd>Optional list of initial mean vectors.</dd>
<dt>fix_means</dt>
<dd>Fix means to their starting values? Must initialize.</dd>
<dt>init_covs</dt>
<dd>Optional list of initial covariance matrices.</dd>
<dt>init_props</dt>
<dd>Optional vector of initial cluster proportions.</dd>
<dt>maxit</dt>
<dd>Maximum number of EM iterations.</dd>
<dt>eps</dt>
<dd>Minimum acceptable increment in the EM objective.</dd>
<dt>report</dt>
<dd>Report fitting progress?</dd></dl>

Arguments

Fit Multivariate Mixture Distribution — FitMix

<dl>

<dt>data</dt>
<dd>Numeric data matrix.</dd>


<dt>k</dt>
<dd>Number of mixture components. Defaults to 2.</dd>


<dt>init_means</dt>
<dd>Optional list of initial mean vectors.</dd>


<dt>fix_means</dt>
<dd>Fix means to their starting values? Must initialize.</dd>


<dt>init_covs</dt>
<dd>Optional list of initial covariance matrices.</dd>


<dt>init_props</dt>
<dd>Optional vector of initial cluster proportions.</dd>


<dt>maxit</dt>
<dd>Maximum number of EM iterations.</dd>


<dt>eps</dt>
<dd>Minimum acceptable increment in the EM objective.</dd>


<dt>report</dt>
<dd>Report fitting progress?</dd>

</dl>

FitMix: Fit Multivariate Mixture Distribution

Description

Usage

Value

Arguments