fmx_normix {QuantileGH}

Naive Parameter Estimates using Mixture of Normal

Description

Naive parameter estimates for finite mixture distribution fmx using mixture of normal distributions.

Usage

fmx_normix(x, K, distname = c("norm", "GH", "sn"), alpha = 0.05, R = 10L, ...)

Arguments

x

numeric vector, observations

K

integer scalar, number of mixture components

distname

character scalar, name of parametric distribution of the mixture components

alpha

numeric scalar, proportion of observations to be trimmed in trimmed k-means algorithm tkmeans

R

integer scalar, number of normalmixEM replicates

...

additional parameters, currently not in use

Details

fmx_normix ... the cluster centers are provided as the starting values of \mu's for the univariate normal mixture by EM algorithm. R replicates of normal mixture estimates are obtained, and the one with maximum likelihood will be selected

Value

Function fmx_normix returns an fmx object.