mixOnekn {MixSemiRob}R Documentation

Two-component Normal Mixture Estimation with One Known Component

Description

‘mixOnekn’ is used for the estimation of the following two-component mixture model:

h(x;θ,f)=πf(xμ1)+(1π)f(xμ2),h(x;\boldsymbol{\theta},f) = \pi f(x-\mu_1)+(1-\pi)f(x-\mu_2),

where θ=(π,μ1,μ2)\boldsymbol{\theta}=(\pi,\mu_1,\mu_2)^{\top} is the parameter to estimate, ff is an unknown density function that is symmetric at zero. The parameters are estimated by assuming ff is the normal density and the first component has a mean of 0. This function can be used to obtain initial values for the mixMPHD function.

Usage

mixOnekn(x, sigma.known = NULL)

Arguments

x

a vector of observations.

sigma.known

standard deviation of the first component (if known). Default is NULL, which calculates the component standard deviations using the given observations x.

Value

A list containing the following elements:

mu

estimated 2 component means, where the first mean is 0.

sigma

estimated 2 component standard deviations.

pi

estimated 2 mixing proportions.

lik

final likelihood.

See Also

mixMPHD

Examples

# see examples for the `mixMPHD' function.

[Package MixSemiRob version 1.1.0 Index]