sars (version 2.1.1)

logLik.thresholdInt: Extract Log-Likelihood for the thresholdInt class

Description

Extract the log-likelihood of sar_threshold model fits, adjusting the number of parameters to account for the breakpoint search. This is a method function for the logLik generic.

Usage

# S3 method for thresholdInt
logLik(object, ...)

Value

Returns an object of class logLik, where the "df" (degrees of freedom) attribute has been adjusted to account for the breakpoint search. See

logLik for more details on objects of class logLik.

Arguments

object

Threshold model fits of class thresholdInt.

...

Additional arguments.

Author

Thomas J. Matthews

Details

There has been considerable debate regarding the number of parameters that are included in different piecewise models. Here (and thus in our calculation of AIC, AICc, BIC etc) we consider ContOne to have five parameters, ZslopeOne - 4, DiscOne - 6, ContTwo - 7, ZslopeTwo - 6, DiscTwo - 9. This function adjusts the degrees of freedom associated with the log likelihood to match the number of parameters listed above. See Matthews & Rigal (2021) for more information.

References

Matthews, T.J. & Rigal, F. (2021) Thresholds and the species–area relationship: a set of functions for fitting, evaluating and plotting a range of commonly used piecewise models. Frontiers of Biogeography, 13, e49404.

Examples

data(aegean2)
a2 <- aegean2[1:168,]
fitT <- sar_threshold(data = a2, mod = c("ContOne", "DiscOne"), 
interval = 0.1, non_th_models = TRUE, logAxes = "area", logT = log10)
logLik(fitT[[1]][[1]])