regmixCrit: regmixCrit

In hkasahar/normalregMix: normalregMix: An R Package for Normal Mixture Regression Models

Description

Computes the critical values of the modified EM test.

Usage

regmixCrit(y, x, parlist, z = NULL, values = NULL, parallel = 1,
  cl = NULL, nrep = 1000, ninits.crit = 25)

Arguments

y	n by 1 vector of data for y
x	n by q matrix of data for x
parlist	The parameter estimates as a list containing alpha, mu, sigma, and gamma in the form of (alpha = (alpha_1, ..., alpha_m), mu = (mu_1, ..., mu_m), sigma = (sigma_1, ..., sigma_m), gam = (gamma_1, ..., gamma_m))
z	n by p matrix of regressor associated with gamma
values	3 by 1 Vector of length 3 (k = 1, 2, 3) at which the p-values are computed
parallel	Determines whether package doParallel is used for calculation
cl	Cluster used for parallelization; if it is NULL, the system will automatically
nrep	The number of replications used to compute p-values
ninits.crit	The number of initial guesses to form critical values

Value

A list with the following items:

crit	3 by 3 matrix of (0.1, 0.05, 0.01 critical values), jth row corresponding to k=j
pvals	A vector of p-values at k = 1, 2, 3

hkasahar/normalregMix documentation built on May 17, 2019, 4 p.m.