|
Activity Number:
|
137
|
|
Type:
|
Topic Contributed
|
|
Date/Time:
|
Monday, July 30, 2007 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Bayesian Statistical Science
|
| Abstract - #309941 |
|
Title:
|
Objective Bayesian Analysis in Memory Study
|
|
Author(s):
|
Xiaoyan Lin*+ and Dongchu Sun
|
|
Companies:
|
University of Missouri-Columbia and University of Missouri-Columbia
|
|
Address:
|
Statistics Dept., 29 Broadway Village Dr, Columbia, MO, 65201,
|
|
Keywords:
|
Objective Bayisan ; Hierarchical Model ; Right Haar prior ; Two-level Wishart prior ; Gibbs sampler
|
|
Abstract:
|
Psychologists believe memory consists two primary components, "conscious recollection" and "automatic activation." A famous approach to measure the two memory components is the Process Dissociation Procedure. Two sets of generalized linear additive models are used to model the recollection ability and automatic activation ability, respectively. Additive components are the effect of a participant and the effect of an item. Objective Bayesian analysis is conducted to estimate the effects. We put the hierarchical priors on the generalized linear models. At the first stage, two bivariate normal priors are used for the effects of participants and the effects of items, respectively. At the second stage, we use objective priors on the two covariance matrices. The simulation is done to illustrate the goodness of the model. The propriety of the posterior distribution is explored.
|
