Bayesian Spatial-Temporal Modeling of Atlantic Cod Abundance in the Gulf of Maine

• Activity Number: 21
• Type: Topic Contributed
• Date/Time: Sunday, August 4, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Section on Bayesian Statistical Science
• Author(s): Xia Wang*+ and Ming-Hui Chen and Dipak K. Dey and Chiu-Yen Kou
• Companies: University of Cincinnati and University of Connecticut and University of Connecticut and University of Connecticut
• Keywords: Bayesian Hierarchical ; Dynamic Factor Model ; Predictive Process ; Spatial-temproal model ; Zero-inflated Poisson

Abstract

This study aims at improving estimation and prediction of Atlantic Cod abundance in the Gulf of Maine using spatial-temporal models for zero inflated count data. We employ the Bayesian spatial dynamic factor model framework as proposed in Esther et al. (2011), which allows both flexibility in the model and dimension reduction. The proposed model not only models the spatial and temporal correlations in the variation in the abundance but also deals with the challenges posed by zero counts, uneven sampling intensities and inference on missing spots. The proposed model is employed in the analysis of the survey data by the Northeast Fisheries Sciences Center (NEFSC). Model comparisons show that the estimation and prediction is improved by the proposed spatial-temporal model.