N-mixture models from a Bayesian perspective (Fabian Ketwaroo)

16:00, February 28, Kennedy Seminar Room 2

Abstract:  N-mixture models are a class of hierarchical models that are commonly used to estimate the absolute abundance of a species based on survey sampling.  This talk will introduce well known N-mixture models as well as newly developed N-mixture models.  We then move on to discuss an analysis of real count data using these N-mixture models. Finally,  we will discuss the identifiability of N-mixture models based on  Wantanable-Akaike-information criterion (WAIC) along with a test for detection probability identifiability.