|
Activity Number:
|
235
- Spatio-Temporal Theory and Methods
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, July 31, 2017 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
|
Abstract #323341
|
View Presentation
|
|
Title:
|
Block Conditioning Approximations Based on Hierarchical Decompositions for High-Dimensional Multivariate Normal Probabilities
|
|
Author(s):
|
Jian Cao* and Marc G. Genton and David Keyes and George Turkiyyah
|
|
Companies:
|
and KAUST and King Abdullah University of Science and Technology and American University of Beirut
|
|
Keywords:
|
multivariate normal ;
conditioning approximations ;
hierarchical decomposition
|
|
Abstract:
|
We discuss various methods for approximating multivariate normal probabilities based on the hierarchical decomposition of covariance matrix in spatial statistics. Specifically, we compare conditioning formulas based on univariate normal probabilities, bivariate normal probabilities, and multivariate normal probabilities. The latter approach exploits the fact that covariance matrices in spatial statistics can be approximated by matrices with a hierarchical low rank structure. The efficiency from block conditioning approximations and hierarchical matrix manipulations make computing probabilities of thousands of dimensions possible on modern workstations.
|
Authors who are presenting talks have a * after their name.
