ecicControl: Bootstrap distributions of Information Criterion Differences

In mcullan/ecic:

Description

Bootstrap distributions of Information Criterion Differences

Usage

ecicControl(n, true, parameters, best, models, N = 1000, ic = "AIC")

Arguments

n:	Sample size.
true:	A string giving the name of the model to generate data from.
param:	A vector of parameters specifying a fitted true model.
best:	A string giving the name of the model to be given as best.
models:	A string vector of model names for selecting between.
N:	Number of bootstrap samples.
ic:	String giving which information criterion to use (e.g. 'AICc', 'BIC')

Value

An ecicControl object containing the following:

differences:

frequencies: a modelFrequencies object containing bootstrap frequencies, fitted parameters for the bootstrap samples, scores for the bootstrap samples and the samples themselves.

Examples

my.models = c("norm0" "norm1", "norm")
ecicControl(n = 25, true = "norm", param = c(0.3, 1.2),
            my.models, N = 1000, ic = 'AIC')

mcullan/ecic documentation built on Sept. 3, 2019, 9:57 a.m.