Learn R Programming

emax.glm

Expectation Maximization General Linear Modelling Tools in R

Installation

The project is currently being added to CRAN. In the mean time work with the development version:

# install.packages("devtools")
devtools::install_githib("Stat-Cook/emax.glm")

Copy Link

Version

Install

install.packages('emax.glm')

Monthly Downloads

1

Version

0.1.2

License

GPL-3

Maintainer

Robert M. Cook

Last Published

July 4th, 2019

Functions in emax.glm (0.1.2)

make.logLike

Construct a log-likelihood function in the parameters b, for the given link family.
plot.em.glm.summary

Error bar plot of coefficients and errors to inspect class overlap.
summary.em.glm

Summarize EM glm coefficients.
plot.em.glm

Plot fit-parameters and errors
small.em

Carry out several short EM fits to test for optimal starting locations.
emax.glm

General linear regression via Expectation-Maximization.
sim.2

Simulated data set
init.fit

Method to initialize EM parameters. Carries out a single GLM fit and applies random noise to form starting space.
results_simple

Simulated data set
results_k25_n1000_e05

Simulated data set
make_param_errors

Calculate parameter errors via inversion of the Hessian matrix (either pracma or numeric approximations).
em.glm_numeric_fit

Numeric approximation routine
em.glm_pracma_fit

Hessian routine
init.random

Method to initialize EM parameters. Purely standard normal noise.
BIC.em.glm

Calculate the BIC of the em.glm model
sim.3

Simulated data set
residuals.em.glm

Deviance residuals for an 'em.glm' object.
results_k25_n1000

Simulated data set
plot_probabilities.matrix

Plot the class probabilities, both compared to data set index and as histogram.
make.dpois

Build a Poisson log likelihood
make.dbinom

Build a Binomial log likelihood
update_probabilities

Construct normalized class properties for a given set of parameters
summary.small.em

Summarize a small.em class
predict.em.glm

Predict values from an 'em.glm' model.
logLik.em.glm

Calculate log-likelihood of the EM model.
plot_probabilities

Probability plots for the K classes fit
plot_probabilities.em.glm

Test Plot em.glm
sim.1

Simulated data set
select_best

Select the best parameters from a set of results
AIC.em.glm

Calculate the AIC of the em.glm model
IC.em.glm

General Information Criteria function
dprob.list

List of distribution functions accessed by family name ("poisson" or "binomial").
deviance.em.glm

Model deviance (calculated from deviance residuals)
em.fit_numeric

Carry our the Newton-Raphson optimization of the parameters for given weights via numeric approximations,
data.1

Simulated data set
em.fit_pracma

Carry our the Newton-Raphson optimization of the parameters for given weights via the pracma hessian,
dispersion

Pearson-based dispersion measurements of an 'em.glm' model.
em.glm

Expectation Maximization glm.