Bayesian Analysis of Latent Threshold Dynamic Models

Bayesian Analysis of Latent Threshold Dynamic Models

We discuss a general approach to dynamic sparsity modeling in multivariate time series analysis. Time-varying parameters are linked to latent processes that are thresholded to induce zero values adaptively, providing natural mechanisms for dynamic variable inclusion/selection. We discuss Bayesian mo...

Full description

Bibliographic Details
Main Authors: Nakajima, J, West, M
Other Authors: West, Mike
Format: Others
Published: 2013
Online Access:http://hdl.handle.net/10161/6152
id ndltd-DUKE-oai-dukespace.lib.duke.edu-10161-6152
record_format oai_dc
spelling ndltd-DUKE-oai-dukespace.lib.duke.edu-10161-61522016-11-03T03:32:26ZBayesian Analysis of Latent Threshold Dynamic ModelsNakajima, JWest, MWe discuss a general approach to dynamic sparsity modeling in multivariate time series analysis. Time-varying parameters are linked to latent processes that are thresholded to induce zero values adaptively, providing natural mechanisms for dynamic variable inclusion/selection. We discuss Bayesian model specification, analysis and prediction in dynamic regressions, time-varying vector autoregressions, and multivariate volatility models using latent thresholding. Application to a topical macroeconomic time series problem illustrates some of the benefits of the approach in terms of statistical and economic interpretations as well as improved predictions. Supplementary materials for this article are available online. © 2013 Copyright Taylor and Francis Group, LLC.DissertationWest, Mike2013-04-01Journal Article151 - 164Journal of Business and Economic Statistics, 2013, 31 (2), pp. 151 - 1640735-0015http://hdl.handle.net/10161/61521537-2707Journal of Business and Economic Statistics10.1080/07350015.2012.747847
collection NDLTD
format Others
sources NDLTD
description We discuss a general approach to dynamic sparsity modeling in multivariate time series analysis. Time-varying parameters are linked to latent processes that are thresholded to induce zero values adaptively, providing natural mechanisms for dynamic variable inclusion/selection. We discuss Bayesian model specification, analysis and prediction in dynamic regressions, time-varying vector autoregressions, and multivariate volatility models using latent thresholding. Application to a topical macroeconomic time series problem illustrates some of the benefits of the approach in terms of statistical and economic interpretations as well as improved predictions. Supplementary materials for this article are available online. © 2013 Copyright Taylor and Francis Group, LLC. === Dissertation
author2 West, Mike
author_facet West, Mike
Nakajima, J
West, M
author Nakajima, J
West, M
spellingShingle Nakajima, J
West, M
Bayesian Analysis of Latent Threshold Dynamic Models
author_sort Nakajima, J
title Bayesian Analysis of Latent Threshold Dynamic Models
title_short Bayesian Analysis of Latent Threshold Dynamic Models
title_full Bayesian Analysis of Latent Threshold Dynamic Models
title_fullStr Bayesian Analysis of Latent Threshold Dynamic Models
title_full_unstemmed Bayesian Analysis of Latent Threshold Dynamic Models
title_sort bayesian analysis of latent threshold dynamic models
publishDate 2013
url http://hdl.handle.net/10161/6152
work_keys_str_mv AT nakajimaj bayesiananalysisoflatentthresholddynamicmodels
AT westm bayesiananalysisoflatentthresholddynamicmodels
_version_ 1718390785313865728

Similar Items