icfit-class: Class "icfit"
In mombf: Model Selection with Bayesian Methods and Information Criteria

icfit-class

R Documentation

Class "icfit"

Description

Stores the output of the search for the model with best information criterion value, e.g. produced by bestBIC, bestBIC, bestAIC or bestIC. The class extends a list, so all usual methods for lists also work for icfit objects, e.g. accessing elements, retrieving names etc.

Methods are provided to extract coefficients, predictions, confidence intervals and summary information about the best model.

Objects from the Class

icfit objects are automatically created by a call to bestBIC or similar.

Slots

The class extends a list with elements:

topmodel: names of the variables in the top model
topmodel.fit: top model as fitted by glm
models: data frame with the information criterion for all models (when enumeration is feasible) or those visited by an MCMC model search in modelSelection (when enumeration is not feasible)
varnames: the names of all variables in the design matrix
msfit: Output of modelSelection (used to search the top model)

Methods

coef: glm fit for the top model
confint: Confidence intervals under the top model
predict: Predictions for the top model (predict.glm)
show: signature(object = "icfit"): Displays general information about the object.