Efficient Maximum Likelihood Estimation for Fixed-Rank Kriging

• Activity Number: 30
• Type: Contributed
• Date/Time: Sunday, July 29, 2012 : 2:00 PM to 3:50 PM
• Sponsor: Section on Statistical Computing
• Author(s): Karl Pazdernik*+ and Ranjan Maitra and Douglas William Nychka and Stephan Sain
• Companies: and Iowa State University and National Center for Atmospheric Research and Geophysical Statistics Project
• Keywords: fixed rank kriging ; maximum likelihood estimation ; Gaussian random field ; massive data sets

Abstract

In spatial statistics, a common method for prediction over a Gaussian random field (GRF) is maximum likelihood estimation combined with kriging. For massive data sets, kriging is computationally intensive, both in terms of CPU time and memory, and so fixed rank kriging has been proposed as a solution. We develop an alteration to this method by utilizing the approximations made in fixed rank kriging combined with restricted maximum likelihood estimation and sparse matrix methodology. Experiments show that additional gains in computational efficiency can be made without loss of accuracy in prediction. The methodology is extended to climate data archived by the National Climate Data Center.