Akaike's information criterion for FLSAM

Returns Akaike's information criterion for a set of FLSAM objects

AIC(object, ..., k = 2)

Arguments

object

An FLSAM or FLSAMs object

...

Further FLSAMobjects

k

the _penalty_ per parameter to be used; the default 'k = 2' is the classical AIC.

Details

The AIC criteria can be used as a basis for comparing model fits in situations where a likelihood ratio test is not appropriate (i.e. the models are not nested). For more information see the help for AIC in the stats package.

Value

Numeric value

Author

Based on code by Anders Nielsen in SAM. Adapted to FLSAM by Mark R. Payne.

See also

lr.test, AIC

Examples

#Load assessment
library(FLSAM)
data(NSH.sam)
#Extract AIC
AIC(NSH.sam)
#> [1] 372.8895
#> attr(,"logarithm")
#> [1] TRUE
#For more an FLSAMs
#data(HERAS.sams)
#plot(AIC(HERAS.sams),pch=16)