MultiBatchList: Constructor for a list of MultiBatch objects
In scristia/CNPBayes: Bayesian mixture models for copy number polymorphisms

Description Usage Arguments Value See Also

MultiBatch is the container used by CNPBayes for the organization of the primary data, current values of the mixture model parameters, hyperparameters, and MCMC chains. A list of MultiBatch objects is often created internally to evaluate and compare different models.

MultiBatchList(
  models,
  data = modelData(),
  K,
  specs = modelSpecs(models, K, data),
  num_models = nrow(specs),
  burnin = 1000L,
  iter = 1000L,
  thin = 1L,
  nStarts = 4L,
  max_burnin = 32000L,
  hp = Hyperparameters(),
  mp = McmcParams(iter = iter, thin = thin, nStarts = nStarts, burnin = burnin,
    max_burnin = max_burnin),
  parameters = modelParameters(hp = hp, mp = mp),
  chains = listChains2(specs, parameters),
  current_values,
  summaries = listSummaries(specs),
  flags = listFlags(specs)
)

`models`	vector of abbreviated model names
`data`	one-dimensional summaries of log-ratios at a CNV-region for a collection of samples. See details for required columns.
`K`	vector for numer of mixture components in the list object
`specs`	additional model specifications
`num_models`	ignored
`burnin`	number of burnin MCMC simulations
`iter`	total number of saved MCMC iterations (after thinning and does not include burnin)
`thin`	number indicating how often MCMC updates are saved in the McmcChains slot of a MultiBatch instance
`nStarts`	number of independent chains
`max_burnin`	ignored
`hp`	a Hyperparameters instance
`mp`	a McmcParams instance
`parameters`	Parameters of the finite Bayesian mixture model
`chains`	a McmcChains instance
`current_values`	values of mixture model parameters from the last iteration of the MCMC. These values can be used to initialize another chain if more MCMC simulations are needed.
`summaries`	list of model summaries
`flags`	list of flags that could indicate possible problems with convergence