glmulti-methods: Methods for Function glmulti: different ways to call glmulti

Description Methods

Description

codeglmulti finds what are the n best models (the confidence set of models) among all possible models (the candidate set, as specified by the user). Models are fitted with the specified fitting function (default is glm) and are ranked with the specified Information Criterion (default is aic). The best models are found either through exhaustive screening of the candidates or using a genetic algorithm, which allows very large candidate sets to be adressed. The output can be used for model selection, variable selection, and multimodel inference.

Methods

y = "ANY", xr = "ANY", data = "ANY", exclude = "ANY"

This will stop with a warning that an improper call has been attempted.

y = "character", xr = "character", data = "ANY", exclude = "ANY"

Calling glmulti with the names of the response variable and of the predictors as character strings. This is the original interface used in versions earlier than 0.6-1.

y = "character", xr = "missing", data = "ANY", exclude = "missing"

Calling glmulti with a model formula represented as a character string. E.g. "u~c+x"

y = "formula", xr = "missing", data = "ANY", exclude = "missing"

Calling glmulti with a model formula containing all the terms to be included in candidate models. E.g. u~c+x

y = "list", xr = "ANY", data = "ANY", exclude = "ANY"

Calling glmulti on a list of (fitted) model objects. Models will not be refitted, but the information criteria will be computed and a regular glmulti object is returned.

y = "glm", xr = "missing", data = "ANY", exclude = "missing"

Calling glmulti with a glm object from which the formula and other parameters will be extracted.

y = "lm", xr = "missing", data = "ANY", exclude = "missing"

Calling glmulti on a lm object from which the formula and other parameters will be extracted.

y = "missing", xr = "ANY", data = "ANY", exclude = "ANY"

If y is missing the glmulti version currently used is printed.


glmulti documentation built on July 2, 2020, 3:19 a.m.