This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.

Abstract Details

Activity Number: 576
Type: Topic Contributed
Date/Time: Wednesday, August 4, 2010 : 2:00 PM to 3:50 PM
Sponsor: SSC
Abstract - #308399
Title: Reversible Jump MCMC for Parametric Functional Data Models: Changing Models Through Function Space Mappings
Author(s): David A. Campbell*+
Companies: Simon Fraser University
Address: 13450 102nd avenue, Surrey, BC, V3T0A3, Canada
Keywords: Reversible Jump MCMC ; Functional Data Analysis ; Differential Equation Models ; Model Selection
Abstract:

Computing the posterior probability of a model from a set of models, MCMC must be able to visit all potential models. Sampling values from different models by Reversible Jump MCMC (RJMCMC) requires altering the dimension of the Markov chain. The main difficulty with RJMCMC is determining a transformation that changes the parameter dimension and proposes parameters into an informative location of a new model. When the models in question are non-nested differential equation systems, this is further complicated by the variety of dynamics that can be produced by a model. I propose a new algorithm to improve the model changing and dimension altering process by stepping through the function spaces of the dynamic systems. The proposed parameter dimension is augmented or reduced through a direction orthogonal to the function space of the previous model but in the space of the proposed model.


The address information is for the authors that have a + after their name.
Authors who are presenting talks have a * after their name.

Back to the full JSM 2010 program




2010 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.