A Test for Stationarity for Irregularly Spaced Spatial Random Field

• Activity Number: 558
• Type: Contributed
• Date/Time: Wednesday, August 6, 2014 : 2:00 PM to 3:50 PM
• Sponsor: IMS
• Author(s): Soutir Bandyopadhyay*+ and Suhasini Subba Rao
• Companies: Lehigh University and Texas A&M
• Keywords: stationarity ; DFT

Abstract

The analysis of spatial data is based on a set of assumptions, which in practice need to be checked. A commonly used assumption is that the spatial random field is second order stationary. In this paper, a test for spatial stationarity for irregular sampled data is proposed. The test is based on a transformation of the data (a type of Fourier transform), where the correlations between the transformed data is close to uncorrelated if the random field is second order stationary, whereas this property does not hold if the random field is second order nonstationary. Using this property a test for second order stationarity is constructed, which is based on measure the degree of correlation in the transformed data. The asymptotic sampling properties of the test statistic is derived under both stationarity and nonstationarity and the method is illustrated with simulations.