Does non-stationary spatial data always require non-stationary random fields?
Does non-stationary spatial data always require non-stationary random fields?
复制标题
DOI：
10.1016/j.spasta.2015.10.001
复制
发表时间：
2015-11-01
影响因子：
2.3
通讯作者：
Rue, Harard
中科院分区：
文献类型：
--
作者：
Fuglstad, Geir-Arne;Simpson, Daniel;Rue, Harard
平稳空间模型是一种理想化，我们期望物理现象的真正依赖结构在空间上变化，但在实践中我们应该如何处理这种非平稳性？我们研究了将灵活的非平稳模型应用于美国本土年降水量数据集所面临的挑战，其中探索性数据分析显示了非平稳协方差结构的有力证据。本文的目的是在空间数据中检测到非平稳性后研究建模流程。我们表明，存在过度拟合模型的真正危险，并且需要仔细建模才能正确考虑变化的二阶结构。事实上，该示例表明有时不需要非平稳高斯随机场来对非平稳空间数据进行建模。 (C) 2015 Elsevier B.V. 保留所有权利。
A stationary spatial model is an idealization and we expect that the true dependence structures of physical phenomena are spatially varying, but how should we handle this non-stationarity in practice? Westudy the challenges involved in applying a flexible non-stationary model to a dataset of annual precipitation in the conterminous US, where exploratory data analysis shows strong evidence of a non-stationary covariance structure.The aim of this paper is to investigate the modelling pipeline once non-stationarity has been detected in spatial data. We show that there is a real danger of over-fitting the model and that careful modelling is necessary in order to properly account for varying second-order structure. In fact, the example shows that sometimes non-stationary Gaussian random fields are not necessary to model non-stationary spatial data. (C) 2015 Elsevier B.V. All rights reserved.