Nonparametric Estimation of Space-Time Covariance Function

• Activity Number: 52
• Type: Invited
• Date/Time: Sunday, July 29, 2012 : 4:00 PM to 5:50 PM
• Sponsor: Section on Statistics and the Environment
• Author(s): Bo Li*+ and InKyung Choi and Xiao Wang
• Companies: Purdue University and Purdue University and Purdue University
• Keywords: Completely monotone function ; Nonparametric ; Space-time covariance model ; Spline regression

Abstract

Covariance structure modeling plays a key role in the space-time data analysis. Various parametric models have been developed to accommodate the idiosyncratic features of a given data set. However, the parametric models may impose unjustified restrictions to the covariance structure and the procedure of choosing a specific model is often ad-hoc. We propose a nonparametric covariance estimator based on the class of space-time covariance models developed by Gneiting (2002) to avoid the choice of parametric forms. Our estimator is obtained via a nonparametric approximation of completely monotone functions. It is easy to implement and our simulation shows it outperforms the parametric models when there is no clear information on model specification. A comprehensive comparison between the nonparametric estimator and parametric models are illustrated using the Irish wind data.