Title | Bayesian estimation of semiparametric nonlinear dynamic factor analysis models using the Dirichlet process prior. |
Publication Type | Journal Article |
Year of Publication | 2011 |
Authors | Chow, Sy-Miin, Niansheng Tang, Ying Yuan, Xinyuan Song, and Hongtu Zhu |
Journal | Br J Math Stat Psychol |
Volume | 64 |
Issue | Pt 1 |
Pagination | 69-106 |
Date Published | 2011 Feb |
ISSN | 0007-1102 |
Keywords | Affect, Bayes Theorem, Factor Analysis, Statistical, Female, Humans, Individuality, Infant, Newborn, Male, Nonlinear Dynamics, Probability, Psychology, Psychometrics, Statistics, Nonparametric, Stochastic Processes, Young Adult |
Abstract | Parameters in time series and other dynamic models often show complex range restrictions and their distributions may deviate substantially from multivariate normal or other standard parametric distributions. We use the truncated Dirichlet process (DP) as a non-parametric prior for such dynamic parameters in a novel nonlinear Bayesian dynamic factor analysis model. This is equivalent to specifying the prior distribution to be a mixture distribution composed of an unknown number of discrete point masses (or clusters). The stick-breaking prior and the blocked Gibbs sampler are used to enable efficient simulation of posterior samples. Using a series of empirical and simulation examples, we illustrate the flexibility of the proposed approach in approximating distributions of very diverse shapes. |
DOI | 10.1348/000711010X497262 |
Alternate Journal | Br J Math Stat Psychol |
Original Publication | Bayesian estimation of semiparametric nonlinear dynamic factor analysis models using the Dirichlet process prior. |
PubMed ID | 21506946 |
PubMed Central ID | PMC3199348 |
Grant List | AG033387 / AG / NIA NIH HHS / United States P01CA142538-01 / CA / NCI NIH HHS / United States R01 MH086633 / MH / NIMH NIH HHS / United States P01 CA142538 / CA / NCI NIH HHS / United States MH086633 / MH / NIMH NIH HHS / United States UL1 RR025747 / RR / NCRR NIH HHS / United States UL1-RR025747-01 / RR / NCRR NIH HHS / United States R21 AG033387 / AG / NIA NIH HHS / United States |
Bayesian estimation of semiparametric nonlinear dynamic factor analysis models using the Dirichlet process prior.
Project: