em.gauss.opti.groups: Finding the optimal number of components

In sp2019-antibiotics/emGauss: Estimation of ECOFF Value by Mixing Gaussians

Description

This function provides the user information of the fitted distributions in order to choose the optimal number of components.

Usage

em.gauss.opti.groups(y, k, alpha, beta, method = "quantile",
  epsilon = 1e-06)

Arguments

y	A data vector with observed values per bin.
k	maximum numbers of components
alpha	inverse gamma shape parameter
beta	inverse gamma rate parameter
method	method how startvalues should be evaluated. For more details see function CreateCluster
epsilon	stopping criterion

Details

This function fits for each number of components (1:k) a mixing distribution of Gaussians by using the function em.gauss. Furthermore, the fit of the distributions is measured by the two information criteria AIC and BIC.

Value

A list with mu, var, pi, loglik, ecoff, AIC, BIC for each fitted distribution

Examples

y <- c(2, 4, 5,6,5,2,2, 1, 1, 2,  2, 1,6,7,8,7,6, 5, 2,1)
em.gauss.opti.groups(y,
                     k= 2,
                     alpha = 1,
                     beta = 2)

sp2019-antibiotics/emGauss documentation built on Nov. 5, 2019, 9:14 a.m.