UEM {UEM}R Documentation

Estimating and updating mixtures via EM

Description

Tools for Gaussian and Poisson mixtures: Estimating, updating, and k-boxplots.

Details

The DESCRIPTION file:

Package: UEM
Type: Package
Title: Estimating and updating mixtures via EM
Version: 0.3-1
Date: 2020-06-30
Author: Jochen Einbeck, Daniel Bonetti, and Najla Qarmalah
Maintainer: Jochen Einbeck <jochen.einbeck@durham.ac.uk>
Depends: R (>= 3.5.0)
Imports: mvtnorm (>= 1.0-9)
Suggests: qcc, npmlreg
Description: Tools for Gaussian and Poisson mixtures: Estimating, updating, and k-boxplots.
License: GPL (>=2)

Index of help topics:

                        Estimating and updating mixtures via EM
EM                      Fitting and updating mixtures via the EM
                        algorithm
draw1d.umix             Auxiliary drawing functions
energy1                 Energy use data
estep                   E-step and M-step
gaussSim                Simulating from mixture models
init                    Initializing the EM algorithm
kboxplot                k-boxplots for mixture data
logLike                 Log-likelihood of fitted model
plot.umix               Plotting function for 'umix' objects
wtSigma                 Auxiliary functions

The package bundles two pieces of software resulting from publications as listed below;

1. Functions EM, UEM, and updateEM to fit Gaussian and Poisson mixtures as a whole, or sequentially as an update algorithm

2. A function k-boxplot to produce and display k-boxplots for mixture data

Author(s)

Jochen Einbeck, Daniel Bonetti, and Najla Qarmalah

Maintainer: Jochen Einbeck <jochen.einbeck@durham.ac.uk>

References

Einbeck, Jochen & Bonetti, Daniel (2014), A study of online and blockwise updating of the EM algorithm for Gaussian mixtures, in Kneib, Thomas, Sobotka, Fabian, Fahrenholz, Jan & Irmer, Henriette eds, Proceedings of the 29th International Workshop on Statistical Modelling. Goettingen, Germany, 14-18 July 2014 II: 29th International Workshop on Statistical Modelling. Goettingen, University of Goettingen, 35-38.

Qarmalah, Najla M., Einbeck, Jochen & Coolen, Frank P.A. (2018). k-Boxplots for mixture data. Statistical Papers 59(2): 513-528.


[Package UEM version 0.3-1 Index]