B1066
Title: A discretized projection-based method for modeling high-dimensional zero-inflated spatial data
Authors: Seiyon Lee - George Mason University (United States) [presenting]
Murali Haran - The Pennsylvania State University (United States)
Abstract: Applications of spatial observations with excessive zeros occur in many disciplines. Modeling such zero-inflated spatial data is computationally challenging, especially in high dimensions. The computational challenge is borne out of inferring the high-dimensional spatial random effects and matrix operations on dense covariance matrices. Markov chain Monte Carlo (MCMC) algorithms may be slow mixing for these models. We propose a computationally efficient approach to model high-dimensional zero-inflated spatial observations using a discretized projections-based approach. Our approach improves mixing in MCMC algorithms and considerably decreases computational overhead for fitting these models. Through simulated examples, we show that our approach performs well in inference and prediction. We also apply our approach to real-world examples in ecology and glaciology.
