| Parameter-class {Umacs} | R Documentation |
This class is used only internally, in Umacs.
This is the base class for all model parameters that need to be
updated within the sampler loop.
Parameter is the parent class for ``sampling schemes" (SamplingScheme-class,
which contain such as Gibbs and Metropolis),
and ``raw code" (RawCode-class).
id:"numeric") an unique id name:"character") full specification of the parameter name, with brackets and but without suffixes. stemname:"character") `stem' of the name, not including brackets or suffixes, (e.g. ``theta"). basename:"character") the name of the parameter, including brackets, but not including suffixes such as ``.mis" (e.g. "theta[1]") savename:"character") the parameter specification as the user specified it, (e.g. "theta[1].mis") impute:"logical") (not used) length:"numeric") length of the actual 'missing' values: code{length(theta.mis)} total.length:"numeric") length of the 'stemname' vector : code{length(theta)} dimension:"numeric") check.ok:"logical") code{TRUE} if parameter check went ok vector.names:"character") (not used) value:"numeric") missing:"numeric") snippets:"snippetlist" ~~ Jouni Kerman <jouni@kerman.com>
Kerman, Jouni. Umacs: A Universal Markov Chain Sampler. Technical report, Columbia University, New York.
Umacs-package for an overview of the Umacs package.
Sampler for how to create a sampler function; see the vignette for details.
Package `rv'.