QAIC.RdCalculate a modification of Akaike's Information Criterion for overdispersed count data (or its version corrected for small sample, quasi-), for one or several fitted model objects.
QAIC(object, ..., chat, k = 2, REML = NULL)
QAICc(object, ..., chat, k = 2, REML = NULL)a fitted model object.
optionally, more fitted model objects.
\(\hat{c}\), the variance inflation factor.
the ‘penalty’ per parameter.
optional logical value, passed to the logLik method
indicating whether the restricted log-likelihood or log-likelihood should be
used. The default is to use the method used for model estimation.
If only one object is provided, returns a numeric value with the
corresponding or ; otherwise returns a
data.frame with rows corresponding to the objects.
\(\hat{c}\) is the dispersion parameter estimated from the global model, and can be calculated by dividing model's deviance by the number of residual degrees of freedom.
In calculation of , the number of model parameters is increased by 1 to account for estimating the overdispersion parameter. Without overdispersion, \(\hat{c} = 1\) and is equal to .
Note that glm does not compute maximum-likelihood estimates in models
within the quasi- family. In case it is justified, it can be worked
around by ‘borrowing’ the aic element from the corresponding
‘non-quasi’ family (see ‘Example’).
Consider using negative binomial family with overdispersed count data.
AICc, quasistats:family family used for models with over-dispersion.
Tests for overdispersion in GLM[M]: check_overdispersionperformance.