Abstract Details
Activity Number:
|
666
|
Type:
|
Invited
|
Date/Time:
|
Thursday, August 8, 2013 : 10:30 AM to 12:20 PM
|
Sponsor:
|
Section on Statistics and the Environment
|
Abstract - #307401 |
Title:
|
Time-Structured PCA for Describing Modes of Variability in Large Global Climate Data Sets
|
Author(s):
|
Cari Kaufman*+ and Ben Shaby
|
Companies:
|
UC Berkeley and UC - Berkeley
|
Keywords:
|
PCA ;
Bayesian statistics ;
climate
|
Abstract:
|
Principal components analysis (PCA) is widely used in atmospheric science under the moniker "Empirical Orthogonal Functions" (EOF). The results of this analysis are used to study the primary modes of variability in atmospheric data, which is often high-dimensional in space and time. PCA may be cast in a probabilistic framework and inference carried out from a Bayesian perspective (see e.g. Tipping and Bishop, 1999, JRSSB). We adapt this framework to take into account the spatial and temporal correlation we anticipate in atmospheric data. In addition to incorporating spatial dependence, we describe how to choose prior distributions for the temporal components of the model to isolate various timescales, using an autocorrelation function that includes a periodic component. The model can be used for exploratory purposes by pre-specifying time-scales of interest and examining the corresponding spatial patterns.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2013 program
|
2013 JSM Online Program Home
For information, contact jsm@amstat.org or phone (888) 231-3473.
If you have questions about the Continuing Education program, please 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.
Copyright © American Statistical Association.