A stabilized and versatile spatial prediction method for geostatistical models

Chun-Shu Chen, Hong Ding Yang, Yang Li

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Geostatistical models are often used to predict spatial variables of interest, but the parameters involved in the spatial correlation function usually cannot be well estimated even if the sample size is large enough under a fixed domain. This would result in the spatial predictor being inaccurate and unstable. In this paper, we apply a data perturbation technique to obtain a new spatial predictor, which is not only continuous but also differentiable with respect to the response variables even after plugging-in the estimated model parameters. Therefore, it is more stable. Moreover, it is known that different spatial predictors obtained from different methods generally have different levels of performance under different circumstances. To mitigate the uncertainties inherent in a selection process, in this paper, we propose an estimation method of weights based on Stein's unbiased risk estimate to combine candidate stabilized spatial predictors, leading to a versatile spatial predictor that is adaptive to the underlying spatial process. Validity for the proposed spatial prediction method is justified both numerically and theoretically. Furthermore, an application of a real data set for the precipitation levels in the Colorado State is also presented.

Original languageEnglish
Pages (from-to)127-141
Number of pages15
JournalEnvironmetrics
Volume25
Issue number2
DOIs
Publication statusPublished - 2014 Jan 1

Fingerprint

Spatial Prediction
Predictors
estimation method
prediction
perturbation
Model
Data Perturbation
Spatial Process
Perturbation Technique
Spatial Correlation
Inaccurate
Differentiable
parameter
method
Correlation Function
Sample Size
Unstable
Uncertainty
Predict
Estimate

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Ecological Modelling

Cite this

Chen, Chun-Shu ; Yang, Hong Ding ; Li, Yang. / A stabilized and versatile spatial prediction method for geostatistical models. In: Environmetrics. 2014 ; Vol. 25, No. 2. pp. 127-141.
@article{8d9a878c12b648b0a8d052f7804c0395,
title = "A stabilized and versatile spatial prediction method for geostatistical models",
abstract = "Geostatistical models are often used to predict spatial variables of interest, but the parameters involved in the spatial correlation function usually cannot be well estimated even if the sample size is large enough under a fixed domain. This would result in the spatial predictor being inaccurate and unstable. In this paper, we apply a data perturbation technique to obtain a new spatial predictor, which is not only continuous but also differentiable with respect to the response variables even after plugging-in the estimated model parameters. Therefore, it is more stable. Moreover, it is known that different spatial predictors obtained from different methods generally have different levels of performance under different circumstances. To mitigate the uncertainties inherent in a selection process, in this paper, we propose an estimation method of weights based on Stein's unbiased risk estimate to combine candidate stabilized spatial predictors, leading to a versatile spatial predictor that is adaptive to the underlying spatial process. Validity for the proposed spatial prediction method is justified both numerically and theoretically. Furthermore, an application of a real data set for the precipitation levels in the Colorado State is also presented.",
author = "Chun-Shu Chen and Yang, {Hong Ding} and Yang Li",
year = "2014",
month = "1",
day = "1",
doi = "10.1002/env.2263",
language = "English",
volume = "25",
pages = "127--141",
journal = "Environmetrics",
issn = "1180-4009",
publisher = "John Wiley and Sons Ltd",
number = "2",

}

A stabilized and versatile spatial prediction method for geostatistical models. / Chen, Chun-Shu; Yang, Hong Ding; Li, Yang.

In: Environmetrics, Vol. 25, No. 2, 01.01.2014, p. 127-141.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A stabilized and versatile spatial prediction method for geostatistical models

AU - Chen, Chun-Shu

AU - Yang, Hong Ding

AU - Li, Yang

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Geostatistical models are often used to predict spatial variables of interest, but the parameters involved in the spatial correlation function usually cannot be well estimated even if the sample size is large enough under a fixed domain. This would result in the spatial predictor being inaccurate and unstable. In this paper, we apply a data perturbation technique to obtain a new spatial predictor, which is not only continuous but also differentiable with respect to the response variables even after plugging-in the estimated model parameters. Therefore, it is more stable. Moreover, it is known that different spatial predictors obtained from different methods generally have different levels of performance under different circumstances. To mitigate the uncertainties inherent in a selection process, in this paper, we propose an estimation method of weights based on Stein's unbiased risk estimate to combine candidate stabilized spatial predictors, leading to a versatile spatial predictor that is adaptive to the underlying spatial process. Validity for the proposed spatial prediction method is justified both numerically and theoretically. Furthermore, an application of a real data set for the precipitation levels in the Colorado State is also presented.

AB - Geostatistical models are often used to predict spatial variables of interest, but the parameters involved in the spatial correlation function usually cannot be well estimated even if the sample size is large enough under a fixed domain. This would result in the spatial predictor being inaccurate and unstable. In this paper, we apply a data perturbation technique to obtain a new spatial predictor, which is not only continuous but also differentiable with respect to the response variables even after plugging-in the estimated model parameters. Therefore, it is more stable. Moreover, it is known that different spatial predictors obtained from different methods generally have different levels of performance under different circumstances. To mitigate the uncertainties inherent in a selection process, in this paper, we propose an estimation method of weights based on Stein's unbiased risk estimate to combine candidate stabilized spatial predictors, leading to a versatile spatial predictor that is adaptive to the underlying spatial process. Validity for the proposed spatial prediction method is justified both numerically and theoretically. Furthermore, an application of a real data set for the precipitation levels in the Colorado State is also presented.

UR - http://www.scopus.com/inward/record.url?scp=84896313590&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84896313590&partnerID=8YFLogxK

U2 - 10.1002/env.2263

DO - 10.1002/env.2263

M3 - Article

AN - SCOPUS:84896313590

VL - 25

SP - 127

EP - 141

JO - Environmetrics

JF - Environmetrics

SN - 1180-4009

IS - 2

ER -