mixAK (version 2.2)

NMixPredDA: Discriminant analysis based on MCMC output from the mixture model

Description

It performs discriminant analysis based on sampled (re-labeled) MCMC chains from the mixture model fitted with NMixMCMC function. Observations to be discriminated may be censored.

Discrimination is based on posterior predictive probabilities of belonging to (re-labeled) mixture components.

Usage

NMixPredDA(object, y0, y1, censor, inity, info)

Arguments

object

an object of class NMixMCMC

y0

vector, matrix or data frame with observations (or limits of censored-observations) to be clustered. See NMixMCMC for details. If y0 is not given then the function discriminates o

y1

vector, matrix or data frame with upper limits of interval-censored observations (if there are any). See NMixMCMC for details.

censor

vector, matrix or data frame with censoring indicators (if there are any censored observations). See NMixMCMC for details.

inity

optional vector, matrix or data frame with initial values of censored observations (if there are any censored observations)

info

number which specifies frequency used to re-display the iteration counter during the computation.

Value

• A data.frame with columns labeled prob1,..., probp giving posterior predictive probabilities of belonging to each component and a column labeled component giving the index of the component with the highest component probability.

See Also

NMixMCMC, NMixPlugDA.