cm: Handling “`cm`” and “`cmList`” objects

Description

Methods for working with candidate model (“cm”) and candidate model list (“cmList”) objects.

Usage

getNthModel(object, index)
getFirstModel(object)
summaryTable(object, index, ...)
"summary"(object, ...)
"print"(x, ...)
"formula"(x, ...)
"IC"(object)
"summary"(object, ...)
"print"(x, ...)

Arguments

object

a list of candidate models of class "cmList" that was returned from a call to fmi or FindMinIC.

index

an index into the cmList, for instance, using index = 4 in getNthModel will return the 4th best candidate model from the list as ranked by IC.

a candidate model of class "cm"

...

some generic methods such as print and summary can take additional arguments, these are passed in to those methods appropriately.

Value

results: a list of candidate models of class "cm".
data: the data passed in including any changes made for groupedData.
best: the best candidate model
modeltype: the model type provided to the FindMinIC call
call: the call to lm or lme that generated the candidate model
IC: the IC of the model fit
formula: the formula of the model fit

Details

Since FindMinIC returns a potentially large list of candidate models, functions such as getNthModel, getFirstModel, summaryTable and summary can be used to either return specific models or summarize the list of models. Once a particular candidate model (cm) has been extracted, functions such as IC, formula, and summary can be used to further understand that particular model.

Examples

Run this code

     data(iris)

     coly="Sepal.Length"
     fixed="Sepal.Width"
     candidates=c("Species","-1","Sepal.Width:Species")

     results.lm = FindMinIC(coly, candidates, fixed, iris)

     print(summary(results.lm))

     # best model:
     print(summary(getFirstModel(results.lm)))
     print(summaryTable(results.lm, 1))
     print(summary(results.lm$results[[1]]))

     # 4th best model:
     print(summary(getNthModel(results.lm, 4)))
     print(summary(results.lm$results[[4]]))