JSM 2015 Preliminary Program

Online Program Home
My Program

Abstract Details

Activity Number: 89
Type: Invited
Date/Time: Sunday, August 9, 2015 : 8:30 PM to 9:15 PM
Sponsor: Korean International Statistical Society
Abstract #317583
Title: Exploratory Data Analysis and High-Performance Computing
Author(s): Doug Nychka*
Companies: National Center for Atmospheric Research
Keywords: spatial statistics ; EDA ; High performance computing ; Climate
Abstract:

Large climate datasets pose problems for exploratory data analysis because of the length of time it takes to try different ideas. This problem fits into the quip: "If I have to wait too long to get my answer I forget my question." In this talk a combination of efficient statistical algorithms, the flexibility of the R statistical language and the architectures of current supercomputers suggest a strategy for data analysis that can provide rapid analysis for the kind of space/time fields that are typical in climate science. As motivation consider the spatial analysis of daily observed temperature fields for the North America and for the past 30 years (~10000 days). Typically each daily field, at least initially, can be analyzed in parallel by single processors (cores) running R. Part of what makes this feasible are spatial statistics models that exploit sparse matrix methods and allow for a daily field to be computed on a single processor. Another component is being able to manage many independent R sessions in an straight forward manner without being an expert on parallel computation.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2015 program





For program information, contact the JSM Registration Department or phone (888) 231-3473.

For Professional Development information, contact the Education Department.

The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.

2015 JSM Online Program Home