Open Access
2016 A comparison of spatial predictors when datasets could be very large
Jonathan R. Bradley, Noel Cressie, Tao Shi
Statist. Surv. 10: 100-131 (2016). DOI: 10.1214/16-SS115

Abstract

In this article, we review and compare a number of methods of spatial prediction, where each method is viewed as an algorithm that processes spatial data. To demonstrate the breadth of available choices, we consider both traditional and more-recently-introduced spatial predictors. Specifically, in our exposition we review: traditional stationary kriging, smoothing splines, negative-exponential distance-weighting, fixed rank kriging, modified predictive processes, a stochastic partial differential equation approach, and lattice kriging. This comparison is meant to provide a service to practitioners wishing to decide between spatial predictors. Hence, we provide technical material for the unfamiliar, which includes the definition and motivation for each (deterministic and stochastic) spatial predictor. We use a benchmark dataset of $\mathrm{CO}_{2}$ data from NASA’s AIRS instrument to address computational efficiencies that include CPU time and memory usage. Furthermore, the predictive performance of each spatial predictor is assessed empirically using a hold-out subset of the AIRS data.

Citation

Download Citation

Jonathan R. Bradley. Noel Cressie. Tao Shi. "A comparison of spatial predictors when datasets could be very large." Statist. Surv. 10 100 - 131, 2016. https://doi.org/10.1214/16-SS115

Information

Received: 1 October 2014; Published: 2016
First available in Project Euclid: 19 July 2016

zbMATH: 1347.62083
MathSciNet: MR3527662
Digital Object Identifier: 10.1214/16-SS115

Subjects:
Primary: 62H11
Secondary: 62P12

Keywords: Best linear unbiased predictor , GIS , massive data , Model selection , reduced rank statistical models

Rights: Copyright © 2016 The author, under a Creative Commons Attribution License

Vol.10 • 2016
Back to Top