|
Activity Number:
|
395
- Statistical Models for High-Dimensional Computer Output
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Tuesday, July 31, 2018 : 2:00 PM to 3:50 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
|
Abstract #328961
|
Presentation
|
|
Title:
|
Computer Model Calibration of Static Systems Using Sequential Monte Carlo Methods
|
|
Author(s):
|
Murali Haran* and Ben Seiyon Lee and Klaus Keller
|
|
Companies:
|
Penn State University and Penn State University and Penn State University
|
|
Keywords:
|
calibration;
complex computer model;
ice sheets;
sequential Monte Carlo;
climate;
projections
|
|
Abstract:
|
Complex computer models play a prominent role in climate science, particularly in projecting future climate. These models have key parameters that need to be inferred ("calibrated") based on observational data. We describe a sequential Monte Carlo method that is well suited for calibration problems for which standard Markov chain Monte Carlo methods and model emulation approaches are computationally burdensome. The motivating scientific problem for our work is a computer model for projecting the future of the Antarctic ice sheet. Our new methods allow us to consider the effects of assimilating information from the Pliocene era.
|
Authors who are presenting talks have a * after their name.
