control_ECM: Control parameters for the ECM algorithm

In tommasorigon/LSBP: Tractable Bayesian density regression via logit stick-breaking priors

Control parameters for the ECM algorithm

Description

The control_ECM function can be used for specifying the technical settings (i.e. the maximum number of iterations, the tolerance level, and the initialization method), of the LSBP_ECM function.

Usage

control_ECM(maxiter = 10000, tol = 0.001, method_init = "cluster")

Arguments

maxiter	An integer indicating the maximum number of iterations for the LSBP_ECM algorithm.
tol	A real number controlling the convergence of the algorithm. The LSBP_ECM algorithm stops when the difference between consecutive values of the log-posterior is smaller than tol.
method_init	The initialization method. The default method_init='cluster' partitions the covariates using the clara clustering algorithm. Other available options are: method_init='random' and method_init='deterministic'.

Value

The inputs are converted into a list. Missing arguments are filled with default values.

tommasorigon/LSBP documentation built on Feb. 25, 2023, 2:47 a.m.