B0398
Title: A Bayesian semiparametric approach for spatial nonhomogeneous Poisson process with applications
Authors: Jieying Jiao - University of Connecticut (United States) [presenting]
Guanyu Hu - The University of Texas Health Science Center at Houston (United States)
Jun Yan - University of Connecticut (United States)
Abstract: Spatial point pattern data are routinely encountered in various fields such as seismology, ecology, environmental science, and epidemiology. Building a flexible regression model for spatial point process is an important task in order to reveal data's spatial pattern and relationships with various factors. We propose a Bayesian semiparametric regression model for spatial Poisson point process data based on powered Chinese resturant process. Further, we allow variable selection through the spike-slab prior. An efficient Markov chain Monte Carlo (MCMC) algorithm is developed for the proposed methods, followed with an extensive simulation studies to evaluate the empirical performance. The proposed methods are further applied to the analysis of the Forest of Barro Colorado Island (BCI) data.
