model.selection.object: Description of Model Selection Objects

In MuMIn: Multi-Model Inference

Description of Model Selection Objects

Description

An object of class "model.selection" holds a table of model coefficients and ranking statistics. It is produced by dredge or model.sel.

Value

The object is a data.frame with additional attributes. Each row represents one model. The models are ordered by the information criterion value specified by rank (lowest on top).

Data frame columns:

<model terms>	For numeric covariates these columns hold coefficent value, for factors their presence in the model. If the term is not present in a model, value is NA.
<varying arguments>	Optional. If any arguments differ between the modelling function calls (except for formulas and some other arguments), these will be held in additional columns (of class "factor").
df	Number of model parameters
logLik	Log-likelihood (or quasi-likelihood for GEE)
<rank>	Information criterion value
delta	Δ_IC
weight	‘Akaike weights’.

Attributes:

model.calls	A list containing model calls (arranged in the same order as in the table). A model call can be retrieved with getCall(*, i) where i is a vector of model index or name (if given as character string).
global	The global.model object
global.call	Call to the global.model
terms	A character string holding all term names. Attribute "interceptLabel" gives the name of the intercept term.
rank	The rank function used
beta	A character string, representing the coefficient standardizing method used. Either "none", "sd" or "partial.sd"
coefTables	List of matrices of class "coefTable" containing each model's coefficients with std. errors and associated dfs
nobs	Number of observations
warnings	optional (pdredge only). A list of errors and warnings issued by the modelling function during the fitting, with a model number appended to each.

It is not recommended to directly access the attributes. Instead, use extractor functions if possible. These include getCall for retrieving model calls, coefTable and coef for coefficients, and nobs. logLik extracts list of model log-likelihoods (as "logLik" objects), and Weights extracts ‘Akaike weights’.

The object has class c("model.selection", "data.frame").

See Also

dredge, model.sel.

MuMIn documentation built on June 22, 2024, 6:44 p.m.