A S4 class to store control parameters for Weighted Kendall distance model fitting. It is derived from class
RankControlWeightedKendall is derived from virtual class
RankControl. All slots in
RankControl are still valid.
This control class tells the solver to fit a model based on Weighted Kendall distance.
The control parameters that start with prefix
EM_ are intended for the EM iteration. The ones with prefix
SeachPi0 control the behaviour of searching model ranking.
maximum number of EM iteration
convergence error for weights and cluster probabilities in EM iteration
maximum number of iterations in the local search of pi0.
a function object that gives a goodness of fit criterion. The default is log likelihood.
a logical value. If TRUE (by default), immediately traverse to the neighbour if it is better than the current pi0. Otherwise, check all neighbours and traverse to the best one.
a logical value. If TRUE, the location of the current pi0 is shown.
a character string specifying which type of neighbour to use in the local search. Supported values are: "Cayley" to use neighbours in terms of Cayley distance or "Kendall" to use neighbours in terms of Kendall distance. Note that Kendall neighbours are a subset of Cayley neighbours
a list to be passed to
optimx. The list must not contain a component
maximize=TRUEsince internally the negation of the likelihood function is minimized.
A character string specifying which assumption to use when handling top-q rankings. Supported choices are "equal-probability" and "tied-rank".
Want to suggest features or report bugs for rdrr.io? Use the GitHub issue tracker.