The role of the range parameter for estimation and prediction in geostatistics

C. G. Kaufman, B. A. Shaby

Research output: Contribution to journalArticle

25 Citations (Scopus)

Abstract

Two canonical problems in geostatistics are estimating the parameters in a specified family of stochastic process models and predicting the process at new locations. We show that asymptotic results for a Gaussian process over a fixed domain with Matérn covariance function, previously proven only in the case of a fixed range parameter, can be extended to the case of jointly estimating the range and the variance of the process. Moreover, we show that intuition and approximations derived from asymptotics using a fixed range parameter can be problematic when applied to finite samples, even for large sample sizes. In contrast, we show via simulation that performance is improved and asymptotic approximations are applicable for smaller sample sizes when the parameters are jointly estimated. These effects are particularly apparent when the process is mean square differentiable or the effective range of spatial correlation is small.

Original languageEnglish (US)
Pages (from-to)473-484
Number of pages12
JournalBiometrika
Volume100
Issue number2
DOIs
StatePublished - Jun 1 2013

Fingerprint

Geostatistics
geostatistics
Random processes
Sample Size
Stochastic Processes
Intuition
prediction
Prediction
Range of data
stochastic processes
sampling
Covariance Function
Small Sample Size
Asymptotic Approximation
Spatial Correlation
Gaussian Process
Mean Square
Process Model
Differentiable
Stochastic Model

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Mathematics(all)
  • Agricultural and Biological Sciences (miscellaneous)
  • Agricultural and Biological Sciences(all)
  • Statistics, Probability and Uncertainty
  • Applied Mathematics

Cite this

Kaufman, C. G. ; Shaby, B. A. / The role of the range parameter for estimation and prediction in geostatistics. In: Biometrika. 2013 ; Vol. 100, No. 2. pp. 473-484.
@article{8334395a7aae44a88b4845d3026c87dc,
title = "The role of the range parameter for estimation and prediction in geostatistics",
abstract = "Two canonical problems in geostatistics are estimating the parameters in a specified family of stochastic process models and predicting the process at new locations. We show that asymptotic results for a Gaussian process over a fixed domain with Mat{\'e}rn covariance function, previously proven only in the case of a fixed range parameter, can be extended to the case of jointly estimating the range and the variance of the process. Moreover, we show that intuition and approximations derived from asymptotics using a fixed range parameter can be problematic when applied to finite samples, even for large sample sizes. In contrast, we show via simulation that performance is improved and asymptotic approximations are applicable for smaller sample sizes when the parameters are jointly estimated. These effects are particularly apparent when the process is mean square differentiable or the effective range of spatial correlation is small.",
author = "Kaufman, {C. G.} and Shaby, {B. A.}",
year = "2013",
month = "6",
day = "1",
doi = "10.1093/biomet/ass079",
language = "English (US)",
volume = "100",
pages = "473--484",
journal = "Biometrika",
issn = "0006-3444",
publisher = "Oxford University Press",
number = "2",

}

The role of the range parameter for estimation and prediction in geostatistics. / Kaufman, C. G.; Shaby, B. A.

In: Biometrika, Vol. 100, No. 2, 01.06.2013, p. 473-484.

Research output: Contribution to journalArticle

TY - JOUR

T1 - The role of the range parameter for estimation and prediction in geostatistics

AU - Kaufman, C. G.

AU - Shaby, B. A.

PY - 2013/6/1

Y1 - 2013/6/1

N2 - Two canonical problems in geostatistics are estimating the parameters in a specified family of stochastic process models and predicting the process at new locations. We show that asymptotic results for a Gaussian process over a fixed domain with Matérn covariance function, previously proven only in the case of a fixed range parameter, can be extended to the case of jointly estimating the range and the variance of the process. Moreover, we show that intuition and approximations derived from asymptotics using a fixed range parameter can be problematic when applied to finite samples, even for large sample sizes. In contrast, we show via simulation that performance is improved and asymptotic approximations are applicable for smaller sample sizes when the parameters are jointly estimated. These effects are particularly apparent when the process is mean square differentiable or the effective range of spatial correlation is small.

AB - Two canonical problems in geostatistics are estimating the parameters in a specified family of stochastic process models and predicting the process at new locations. We show that asymptotic results for a Gaussian process over a fixed domain with Matérn covariance function, previously proven only in the case of a fixed range parameter, can be extended to the case of jointly estimating the range and the variance of the process. Moreover, we show that intuition and approximations derived from asymptotics using a fixed range parameter can be problematic when applied to finite samples, even for large sample sizes. In contrast, we show via simulation that performance is improved and asymptotic approximations are applicable for smaller sample sizes when the parameters are jointly estimated. These effects are particularly apparent when the process is mean square differentiable or the effective range of spatial correlation is small.

UR - http://www.scopus.com/inward/record.url?scp=84878069118&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84878069118&partnerID=8YFLogxK

U2 - 10.1093/biomet/ass079

DO - 10.1093/biomet/ass079

M3 - Article

AN - SCOPUS:84878069118

VL - 100

SP - 473

EP - 484

JO - Biometrika

JF - Biometrika

SN - 0006-3444

IS - 2

ER -