Skip to contents

LHWR AICc

Source: R/diagnostics.R
AICc.lhwr.Rd

LHWR AICc

Usage

AICc.lhwr(object)

AICc.hgwr(object)

Arguments

object

An LHWR model object returned by lhwr() or a low-level fit returned by hgwr.fit().

Value

A single numeric value containing the corrected Akaike information criterion. Smaller values indicate a better fitted model under this criterion.

Details

This extractor does not refit or select a model. AICc requires hat-matrix diagnostics; a nonpositive n - tr(S) - 2 denominator produces infinity. Gaussian AIC and BIC are also stored in object$LHWR.diagnostic, but bw.lhwr() directly selects only by AICc or CV. See vignette("lhwr-diagnostics-and-tuning", package = "LHWL") for formulas.