em.glm_numeric_fit: Numeric approximation routine

In emax.glm: General Tools for Building GLM Expectation-Maximization Models

Description

Numeric approximation routine

Usage

em.glm_numeric_fit(x, y, b.list, class_probs, weight = c(1), K = 2,
  tol.1 = 1e-08, debug = TRUE, family = poisson(), maxiter = Inf)

Arguments

x	An n-by-p design matrix.
y	A vector of observation of length n.
b.list	List of K-classes each entry being a k length parameter vector,
class_probs	Matrix (n x K) of normalized class probabilities.
weight	A n length vector of observation weight terms. This is currently designed to be either the exposure for a Poisson model or the number of trials for a Logistic model.
K	Number of EM classes to be fit.
tol.1	Tolerance of the NR minimization.
debug	Boolen flag. Turn on to check optimization steps via messages.
family	GLM family to fit with.
maxiter	Maximum iterations of the NR methods for exiting before convergence.

Value

A list of parameter values on convergence for each of k-classes.

Examples

x <- model.matrix(~ 1 , data = warpbreaks)
y <- warpbreaks$breaks

b.list <- list(1, 1)
class_probs = matrix(rep(0.5, 54*2), ncol = 2)

em.glm_numeric_fit(x = x, y = y, b.list = b.list, class_probs = class_probs)

emax.glm documentation built on July 4, 2019, 5:04 p.m.