Calculates the Hannan-Quinn Information Criterion for objects of class pcrfit, nls, lm, glm or any other models from which logLik, coef and residuals can be extracted. It is somewhat similar to BIC, but penalizes $n$ even more by double logarithmation.
Usage
HQIC(object)
Arguments
object
a fitted model.
Value
The Hannan-Quinn Information Criterion.
Details
$$HQIC = -2 \cdot log(\mathcal{L}_{max}) + 2 \cdot k \cdot log(log(n))$$
with $\mathcal{L}_{max}$ = maximum likelihood, $k$ = number of parameters and $n$ = number of observations.
References
The Determination of the Order of an Autoregression.
Hannan EJ & Quinn BG.
J Roy Stat Soc B (1979), 41: 190-195.