modelsupport {MorphoRegions} | R Documentation |
Evaluate model support
Description
modelsupport()
computes measures of the relative support of each of the best models identified by modelselect()
to facilitate selecting the optimal number and position of regions. These measures are in the form of information criteria (AICc and BIC).
Usage
modelsupport(models)
Arguments
models |
a |
Value
A regions_modelsupport
object, which contains the best model for each number of regions as determined by the AICc and BIC. The computed statistics are AICc
/BIC
–the value of the information criterion (IC) for each model, deltaAIC
/deltaBIC
–the difference between the IC for the corresponding model and that of the model with the lowest IC value, model_lik
–the likelihood ratio of the model against the model with the lowest IC value, and Ak_weight
/BIC_weight
–the Akaike weights for each model used to compute the region score. The region score is a weighted average of the numbers of regions, weighted by the Akaike weights to represent the variability around the optimal number of regions.
See Also
modelselect()
, calcregions()
, calcBPvar()
, modelperf()
, plotsegreg()
Examples
data("alligator")
alligator_data <- process_measurements(alligator,
pos = "Vertebra")
# Compute PCOs
alligator_PCO <- svdPCO(alligator_data)
# Fit segmented regression models for 1 to 7 regions
# using PCOs 1 to 4 and a continuous model with a
# non-exhaustive search
regionresults <- calcregions(alligator_PCO,
scores = 1:4,
noregions = 7,
minvert = 3,
cont = TRUE,
exhaus = FALSE,
verbose = FALSE)
regionresults
# For each number of regions, identify best
# model based on minimizing RSS
bestresults <- modelselect(regionresults)
bestresults
# Evaluate support for each model and rank models
supp <- modelsupport(bestresults)
supp
# 5 regions best based on AICc; 6 regions based on BIC