Abstract #300271

The views expressed here are those of the individual authors
and not necessarily those of the ASA or its board, officers, or staff.

Back to main JSM 2002 Program page
JSM 2002 Abstract #300271
Activity Number: 238
Type: Invited
Date/Time: Tuesday, August 13, 2002 : 2:00 PM to 3:50 PM
Sponsor: Business & Economics Statistics Section*
Abstract - #300271
Title: Continuous-Time Nonlinear Autoregressive Models
Author(s): Osnat Stramer*+
Affiliation(s): University of Iowa
Address: 372 Schaeffer Hall, Iowa City, Iowa, 52242,
Keywords: Autoregressive model ; Diffusion process ; MCMC algorithms
Abstract:

Some procedures are introduced in Tsai & Chan (2000) for checking the appropriateness of fitting continuous-time linear autoregressive (LCAR) models versus continuous-time nonlinear (NLCAR) models. A broad class of NLCAR models that includes the continuous-time threshold models is introduced in Tsai & Chan (2000).

The statistical inference problem for NLCAR processes is more difficult than for LCAR processes, since the likelihood, given a discrete set of observations, is unavailable, and thus the existing estimation methods in the literature are somewhat ad hoc.

This talk will introduce a new MCMC approach to Bayesian analysis of NLCAR models. Obvious ways to implement MCMC on these models leads to algorithms which have extremely poor mixing problems. A method of parameterizing the model that breaks down the correlation structure between the unobserved latent process and its parameters will be introduced.

  • 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 2002 program

JSM 2002

For information, contact meetings@amstat.org or phone (703) 684-1221.

If you have questions about the Continuing Education program, please contact the Education Department.

Revised March 2002