AICtable {wiqid}

Make a table for AIC or other criterion

Description

Takes the output from a call to AIC or AICc returns a data frame with model likelihoods and model weights.

Usage

AICtable(x, digits=3, sort)

Arguments

x

A data frame with the second column being AIC-type values, as returned by AIC or AICc.

digits

integer indicating the number of decimal places to retain.

sort

logical indicating whether the table should be sorted by AIC; if missing, the rows are sorted if the table has row names.

Value

Returns the original data frame with the following extra columns

Delta

The difference between each value in x and min(x)

ModelLik

The model likelihood

ModelWt

The model weight

If sort = TRUE, the rows will be sorted by increasing values of AIC/AICc.

Author(s)

Mike Meredith

References

Burnham and Anderson (2002) Model selection and multimodel inference: a practical information-theoretic approach, 2 edn. Springer-Verlag.

Examples

data(KanhaTigers)
mods <- NULL
mods$M0 <- closedCapM0(KanhaTigers)
mods$Mh2 <- closedCapMh2(KanhaTigers)
mods$MhJK <- closedCapMhJK(KanhaTigers)
mods$Mt <- closedCapMt(KanhaTigers)
AICc <- sapply(mods, AICc)
AICtable(AICc)
# MhJK does not use likelihood maximisation