The views expressed here are those of the individual authors and not necessarily those of the JSM sponsors, their officers, or their staff.
Abstract Details
Activity Number:
|
156
|
Type:
|
Topic Contributed
|
Date/Time:
|
Monday, August 1, 2011 : 10:30 AM to 12:20 PM
|
Sponsor:
|
International Chinese Statistical Association
|
Abstract - #301983 |
Title:
|
Adaptive Smoothing with a Cauchy Process Prior
|
Author(s):
|
Paul Speckman*+
|
Companies:
|
University of Missouri at Columbia
|
Address:
|
134B Middlebush Bldg., Columbia, MO, 65211-6100,
|
Keywords:
|
smoothing spline ;
adaptive estimate ;
Cauchy process
|
Abstract:
|
The equivalence between spline smoothing, which penalizes the $L_2$ norm of a derivative, and Bayesian inference with an integrated Brownian motion prior is well known. In this talk, we explore the use of Cauchy process in place of Gaussian process priors. For simplicity, we use a discrete approximation to the derivative, so that an exact, explicit solution exists. We demonstrate by example that the resulting Bayes estimator has many of the desirable properties of a penalized estimator with an $L_1$ penalty, for example. Moreover, fully Bayesian inference is easily obtained through efficient MCMC techniques. The prior can be extended to provide adaptive smoothing on a lattice in two dimensions. Illustrations with simulated and real data are provided.
|
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 2011 program
|
2011 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.