distrib_diff_lcwm_g {outlierMBC}R Documentation

Compute the dissimilarity for a single component of a Linear CWM.

Description

Computes the covariate dissimilarity value, the response dissimilarity value, and their aggregated dissimilarity value. It also obtains the covariate, response, and joint densities for every observation.

Usage

distrib_diff_lcwm_g(x, z_g, mu_g, sigma_g, mod_g, y_sigma_g, dd_weight = 0.5)

Arguments

x

Covariate data only.

z_g

Component assignment probability vector.

mu_g

Component mean vector for the covariates.

sigma_g

Component covariance matrix for the covariates.

mod_g

Component regression model.

y_sigma_g

Component regression standard deviation for the response.

dd_weight

A value between 0 and 1 which controls the weighting of the response and covariate dissimilarities when aggregating.

Value

distrib_diff_lcwm_lcwm_g returns a list with the following elements:

diff

Aggregated dissimilarity value for this component.

dens

Joint (covariate & response) density of all observations for this component.

diff_x

Covariate dissimilarity value for this component.

diff_y

Response dissimilarity value for this component.

dens_x

Covariate density of all observations for this component.

dens_y

Response density of all observations for this component.


[Package outlierMBC version 0.0.1 Index]