dcFit: Internal function for iterative model fitting with data...

In dclone: Data Cloning and MCMC Tools for Maximum Likelihood Methods

Internal function for iterative model fitting with data cloning

Description

This is the workhorse for dc.fit and dc.parfit.

Usage

.dcFit(data, params, model, inits, n.clones,
    multiply = NULL, unchanged = NULL,
    update = NULL, updatefun = NULL, initsfun = NULL,
    flavour = c("jags", "bugs", "stan"),
    n.chains=3, cl = NULL, parchains = FALSE,
    return.all=FALSE, check.nclones=TRUE, ...)

Arguments

data	A named list (or environment) containing the data.
params	Character vector of parameters to be sampled. It can be a list of 2 vectors, 1st element is used as parameters to monitor, the 2nd is used as parameters to use in calculating the data cloning diagnostics.
model	Character string (name of the model file), a function containing the model, or a custommodel object (see Examples).
inits	Optional specification of initial values in the form of a list or a function (see Initialization at jags.model). If missing, will be treated as NULL and initial values will be generated automatically.
n.clones	An integer vector containing the numbers of clones to use iteratively.
multiply	Numeric or character index for list element(s) in the data argument to be multiplied by the number of clones instead of repetitions.
unchanged	Numeric or character index for list element(s) in the data argument to be left unchanged.
update	Numeric or character index for list element(s) in the data argument that has to be updated by updatefun in each iterations. This usually is for making priors more informative, and enhancing convergence. See Details and Examples.
updatefun	A function to use for updating data[[update]]. It should take an 'mcmc.list' object as 1st argument, 2nd argument can be the number of clones. See Details and Examples.
initsfun	A function to use for generating initial values, inits are updated by the object returned by this function from the second iteration. If initial values are not dependent on the previous iteration, this should be NULL, otherwise, it should take an 'mcmc.list' object as 1st argument, 2nd argument can be the number of clones. This feature is useful if latent nodes are provided in inits so it also requires to be cloned for subsequent iterations. See Details and Examples.
flavour	If "jags", the function jags.fit is called. If "bugs", the function bugs.fit is called. If "stan", the function stan.fit is called.
n.chains	Number of chains to generate.
cl	A cluster object created by makeCluster, or an integer, see parDosa and evalParallelArgument.
parchains	Logical, whether parallel chains should be run.
return.all	Logical. If TRUE, all the MCMC list objects corresponding to the sequence n.clones are returned for further inspection (this only works with partype = "parchains"). Otherwise only the MCMC list corresponding to highest number of clones is returned with summary statistics for the rest.
check.nclones	Logical, whether to check and ensure that values of n.clones are unique and increasing. check.nclones = FALSE means that n.clones is used as is, thus it is possible to supply repeated values but still use the update functionality.
...	Other values supplied to jags.fit, or bugs.fit, depending on the flavour argument.

Value

An object inheriting from the class 'mcmc.list'.

Author(s)

Peter Solymos, implementation is based on many discussions with Khurram Nadeem and Subhash Lele.

See Also

dc.fit, dc.parfit

dclone documentation built on Aug. 8, 2025, 6:46 p.m.