There was weak evidence for a region effect ([DELTA]QAIC = 0.8), although by convention more complicated models are not favored over simpler models that have lower QAIC scores.
of fixed [DELTA]QAIC * effects Depth + [depth.sup.2])(-) + 10 0 substrate + region Depth + [depth.sup.2](-) + substrate 7 11.1 Substrate + region 7 108 Substrate 5 120 Depth + [depth.sup.2](-) + region 6 269.8 Depth + [depth.sup.2](-) 3 279.1 Region 4 362.2 Depth 2 377.8 Model QAIC * Cumulative weight weight Depth + [depth.sup.2])(-) + 1.00 1.00 substrate + region Depth + [depth.sup.2](-) + substrate 0.00 1.00 Substrate + region 0.00 1.00 Substrate 0.00 1.00 Depth + [depth.sup.2](-) + region 0.00 1.00 Depth + [depth.sup.2](-) 0.00 1.00 Region 0.00 1.00 Depth 0.00 1.00 * Akaike's information criterion corrected for overdispersion.
In an information-theoretic context, over-parameterized models would simply be penalized for requiring additional parameters to explain the same amount of variation in the data and therefore would be unlikely to be selected with QAIC.
3); the base model plus a hook main effect received majority support as the best fitting model and models that included hook-user, hook-leader, and hook-species interactions had QAIC values within four units of the best fitting model.
Quasi-Akaike information criterion (QAIC) was used to evaluate model performance, with the lowest value indicating the most parsimonious model.
Seleccion de Modelos.- Para el Bosque Joven despues de 1000 simulaciones solo 298 mostraron una devianza mayor que la observada en el modelo con el valor mas bajo de QAIC
. Despues de corregir los valores por sobredispersion el mejor modelo fue el reducido ([FI]p) esto es, aquel en el que tanto la supervivencia como las probabilidades de captura son constantes (Cuadro 1).
We selected a model, based on the lowest QAIC
value, in which rates of survival varied by time (trapping session) and rates of recapture varied by group (site).
The highest ranked covariate model in 2001, 2004, and 2010 was fraction of the moon illuminated (QAIC model weight = 0.11, 0.17, and 0.23, respectively).
APPENDIX.--Quasi Akaike Information Criterion (QAIC) model selection used to delineate factors associated with calling (CALL) activity of the endangered Houston toad (Bufo houstonensis) in the Lost Pines ecoregion of Texas, with each year analyzed independently.
) for model selection (Burnham & Anderson 2003).
We calculated quasi-AIC (QAIC) values for the models predicting ASRs to determine the best fit model.
In the Garden PMU, three models were found to be significant ([DELTA] QAIC < 2) in predicting pronghorn ASRs (Table 2) and included 69% of the model weights.