priordefine: Advanced constructor for the 'prior' class

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

Description

This constructor defines a data dependent prior with parameters by matching moments. As a consequence it needs as inputs an fdata object and a model object. The prior distributions chosen and the methods how parameters are computed are described in Frühwirth-Schnatter (2006).

Usage

priordefine(
  fdata = fdata(),
  model = model(),
  varargin = NULL,
  prior.wagner = TRUE,
  s = 5
)

Arguments

fdata	An fdata object holding the data. Observations in slot @y must be existent.
model	A model object specifying the finite mixture model.
varargin	NULL or a prior object. This enables the user to pass in an already constructed prior object that gets then completed.
prior.wagner	A logical indicating, if the prior from Wagner (2007) should be used in case of an exponential mixture model.
s	A numeric specifying the standard deviation s for the Metropolis-Hastings proposal.

Value

A fully specified prior object.

References

• Fr\"uwirth-Schnatter, S. (2006), "Finite Mixture and Markov Switching Models"

• Wagner, H. (2007), "Bayesian analysis of mixtures of exponentials", Journal of Applied Mathematics, Statistics and Informatics 3, 165-183

See Also

• prior for the class definition

• prior() for the default constructor of the class

Examples

# Create a Poisson mixture model.
f_model <- model("poisson", par = list(lambda = c(0.3, 0.7)), K = 2)
# Simulate data from the model.
f_data <- simulate(f_model)
# Use the advanced constructor to generate a prior.
f_prior <- priordefine(f_data, f_model)

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