![]() However, my understanding is that a c-hat of less than 3 indicates adequate fit with minor overdispersion. My questions are as follows: For those models with a c-hat of less than 3, is it justifiable to move forward with inferences once I inflate the SEs by a factor of the c-hat? Or does the p-value of 0 render the c-hat score irrelevant, removing any grounds for advancing with inferences for a model with a c-hat of 2.4 over a model with a c-hat of 6? ![]() All of the models yielded gof p-values that are less than 0.05, indicating a rejection of H0 (H0=model fits the data). As suggested elsewhere in this forum, I have applied the goodness of fit procedure (mb.gof.test) to the subset of top-fitting models because the global model has too many covariates for gof to be evaluated. I am looking to calculate/compare goodness of fit of the top-ranked models. I have a set of top-ranked single season occupancy models fit in unmarked - occu() - selected using AIC-based model ranking (AIC ≤ 2) following a dredge procedure that fit all combinations of biologically viable covariates (package: MuMIn).
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |