Heterogeneity pursuit for spatial point pattern with application to tree locations: A Bayesian semiparametric recourse
Heterogeneity pursuit for spatial point pattern with application to tree locations: A Bayesian semiparametric recourse
复制标题
DOI：
10.1002/env.2694
复制
发表时间：
2020-03
期刊：
影响因子：
1.7
通讯作者：
Jieying Jiao;Guanyu Hu;Jun Yan
中科院分区：
文献类型：
--
作者：
Jieying Jiao;Guanyu Hu;Jun Yan
经常会遇到空间点图案数据。对于描述空间点模式和了解潜在风险因素对这种模式的影响，对于潜在强度的灵活回归模型是必不可少的。我们提出了一个贝叶斯半参数回归模型，其中观察到的空间点服从具有强度函数的空间泊松过程，该强度函数利用乘性协变量效应来调整非参数基线强度。基准强度是按件恒定的，事先采用动力中餐馆流程，以防止不必要的大量件。参数回归部分允许通过对回归系数的尖峰板先验来选择变量。提出了一种有效的马尔可夫链蒙特卡罗算法。该方法的性能在广泛的仿真研究中得到验证。在应用于巴拿马中部巴罗科罗拉多岛森林动态研究样地中悬铃木的位置时，空间异质性除了具有空间变化的基线强度的地理测量外，还归因于土壤测量的子集。
Spatial point pattern data are routinely encountered. A flexible regression model for the underlying intensity is essential to characterizing the spatial point pattern and understanding the impacts of potential risk factors on such pattern. We propose a Bayesian semiparametric regression model where the observed spatial points follow a spatial Poisson process with an intensity function which adjusts a nonparametric baseline intensity with multiplicative covariate effects. The baseline intensity is piece‐wise constant, approached with a powered Chinese restaurant process prior which prevents an unnecessarily large number of pieces. The parametric regression part allows for variable selection through the spike‐slab prior on the regression coefficients. An efficient Markov chain Monte Carlo algorithm is developed for the proposed methods. The performance of the methods is validated in an extensive simulation study. In application to the locations of Beilschmiedia pendula trees in the Barro Colorado Island forest dynamics research plot in central Panama, the spatial heterogeneity is attributed to a subset of soil measurements in addition to geographic measurements with a spatially varying baseline intensity.