On the Effect of Preferential Sampling in Spatial Prediction.

On the Effect of Preferential Sampling in Spatial Prediction.
复制标题

DOI:
10.1002/env.2169
复制
发表时间:
2012-11-01
期刊:
影响因子:
1.7
通讯作者:
Holland, David M.
Holland, David M.
中科院分区:
环境科学与生态学3区
文献类型:
--
作者:
Gelfand, Alan E.;Sahu, Sujit K.;Holland, David M.

文献摘要

参考文献

被引文献

相似文献

空间网中采样点的选择往往以实际需求为指导。特别地,许多位置被优先选择以捕获响应的高值,例如,环境监测中的空气污染水平。然后,由于选择性采样,暴露表面的模型估计和预测变得有偏差。由于预测往往是建模的主要工具,我们建议,优先采样的效果更重要的是在所得到的预测表面比参数估计。我们的贡献是提供一个直接的模拟为基础的方法来评估优惠抽样的影响。我们比较了两个预测表面的研究区域,一个起源于一个“操作”的强度驱动监测站点的选择,另一个在完全的空间随机性的概念。我们可以考虑一系列的反应模型。它们可以反映操作强度,引入替代信息协变量,或者只是提出一个灵活的空间模型。然后,我们可以在给定的模型下生成数据。在拟合模型和插值(克里金)后,我们将获得两个预测表面进行比较。重要的是要注意,我们需要合适的度量来比较表面,并且预测表面是随机的,因此我们需要进行预期的比较。
The choice of the sampling locations in a spatial network is often guided by practical demands. In particular, many locations are preferentially chosen to capture high values of a response, for example, air pollution levels in environmental monitoring. Then, model estimation and prediction of the exposure surface become biased due to the selective sampling. Since prediction is often the main utility of the modeling, we suggest that the effect of preferential sampling lies more importantly in the resulting predictive surface than in parameter estimation. Our contribution is to offer a direct simulation-based approach to assessing the effects of preferential sampling. We compare two predictive surfaces over the study region, one originating from the notion of an ‘operating’ intensity driving the selection of monitoring sites, the other under complete spatial randomness. We can consider a range of response models. They may reflect the operating intensity, introduce alternative informative covariates, or just propose a flexible spatial model. Then, we can generate data under the given model. Upon fitting the model and interpolating (kriging), we will obtain two predictive surfaces to compare. It is important to note that we need suitable metrics to compare the surfaces and that the predictive surfaces are random, so we need to make expected comparisons.
DOI: 10.1198/016214507000000031
发表时间: 2007-12-01
影响因子: 3.7
作者:
Sahu, Sujit K.;Gelfand, Alan E.;Holland, David M.
通讯作者: Holland, David M.
DOI: 10.1214/10-aoas335
发表时间: 2010-09-01
影响因子: 1.8
作者:
Chakraborty, Avishek;Gelfand, Alan E.;Silander, John A., Jr.
通讯作者: Silander, John A., Jr.
DOI: 10.1111/j.1467-9868.2008.00663.x
发表时间: 2008-09-01
期刊: Journal of the Royal Statistical Society. Series B, Statistical methodology
影响因子: --
作者:
Banerjee S;Gelfand AE;Finley AO;Sang H
通讯作者: Sang H
DOI: 10.1023/a:1026001008598
发表时间: 2003-01-01
期刊: LANDSCAPE ECOLOGY
影响因子: 5.2
作者:
Hooten, MB;Larsen, DR;Wikle, CK
通讯作者: Wikle, CK
DOI: 10.1111/j.1467-9876.2009.00685.x
发表时间: 2010-01-01
影响因子: 1.6
作者:
Sahu, Sujit K.;Gelfand, Alan E.;Holland, David M.
通讯作者: Holland, David M.