|
Abstract:
|
Habitat suitability modeling methods for presence-only species data are limited in their ability for making true predictions and are therefore often misused in ecological applications. A use-availability design combines presence only species data with a background sample of covariates where the species presence/absence is unknown. Assuming a log link-function for the true probability of presence/absence, the use-availability data then can be analyzed as a logistic regression model with a biased estimate of the intercept. Due to the biased intercept, the model is unable to make true predictions. Instead, ranking the "pseudo-predictions" from the model with biased intercept provides a viable alternative for making predictive inference in single-species models. We show that when such a single species model is extended to multiple species the ranks are no longer conserved across species, limiting predictive inference. Alternatively, by assuming a logit link-function for the true probabilities of the presence/absence data, the resulting generalized nonlinear model allows for predictive inference even when extended to multiple species.
|
