Abstract

Abstract Most data have a space and time label associated with them; data that are close together are usually more correlated than those that are far apart. Prediction (or forecasting) of a process at a particular label where there is no datum, from observed nearby data, is the subject of this article. One approach, known as geostatistics, is featured, from which linear methods of spatial prediction (kriging) will be considered. Brief reference is made to other linear/nonlinear, stochastic/deterministic predictors. The (linear) geostatistical method is applied to piezometric-head data around a potential nuclear-waste repository site.

Keywords

GeostatisticsKrigingVariogramGeodetic datumMathematicsStatisticsComputer scienceData miningGeologySpatial variabilityGeodesy

Affiliated Institutions

Related Publications

Publication Info

Year
1989
Type
article
Volume
43
Issue
4
Pages
197-202
Citations
80
Access
Closed

External Links

Social Impact

Altmetric
PlumX Metrics

Social media, news, blog, policy document mentions

Citation Metrics

80
OpenAlex

Cite This

Noel Cressie (1989). Geostatistics. The American Statistician , 43 (4) , 197-202. https://doi.org/10.1080/00031305.1989.10475658

Identifiers

DOI
10.1080/00031305.1989.10475658