dataclass-class: Finmix 'dataclass' class

In simonsays1980/finmix: An R package for Bayesian estimation of finite mixture distributions

Description

Stores objects to classify observations using a fully specified mixture model. If the indicators are known a finite mixture model is fully specified as then the likelihood can be calculated for each observation depending on the component it stems from.

Slots

logpy

An array containing the logarithmized

prob

An array storing the probability classification matrix that defines for each observation the probability of belonging to component k. Henceforth, each row sums to one. The dimension of this array is N x K.

mixlik

A numeric storing the logarithm of the mixture likelihood evaluated at certain parameters par from a finmix model object and corresponding weights.

entropy

A numeric storing the entropy of the classification.

loglikcd

An array storing the logarithm of the conditional likelihood of each component parameter, if indicators have not been simulated. The array has dimension 1 x K.

postS

A numeric storing the posterior probability of the indicators S in the data, if indicators have been simulated.

References

Frühwirth-Schnatter, S. (2006), "Finite Mixture and Markov Switching Models"

See Also

• fdata for the class holding the data

• model for the class defining a finite mixture model

• dataclass() for the constructor of this class

simonsays1980/finmix documentation built on Dec. 23, 2021, 2:25 a.m.