Discussion on Competition for Spatial Statistics for Large Datasets

TL;DR

This paper reviews the AppStatUZH team's participation in a competition focused on spatial statistical methods for large datasets, highlighting their approaches and results in covariance approximation and prediction.

Contribution

It presents a detailed account of applying covariance tapering and Wendland functions in large-scale spatial data analysis within a competitive framework.

Findings

Effective covariance approximation methods identified

Successful parameter estimation and prediction demonstrated

Comparison of different spatial approximation techniques

Abstract

We discuss the experiences and results of the AppStatUZH team's participation in the comprehensive and unbiased comparison of different spatial approximations conducted in the Competition for Spatial Statistics for Large Datasets. In each of the different sub-competitions, we estimated parameters of the covariance model based on a likelihood function and predicted missing observations with simple kriging. We approximated the covariance model either with covariance tapering or a compactly supported Wendland covariance function.