Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | 김건호 | - |
dc.date.accessioned | 2019-12-10T05:08:26Z | - |
dc.date.available | 2019-12-10T05:08:26Z | - |
dc.date.issued | 2018-11 | - |
dc.identifier.citation | COMPUTATIONAL STATISTICS, v. 33, no. 4, page. 1715-1731 | en_US |
dc.identifier.issn | 0943-4062 | - |
dc.identifier.issn | 1613-9658 | - |
dc.identifier.uri | https://link.springer.com/article/10.1007%2Fs00180-018-0796-9 | - |
dc.identifier.uri | https://repository.hanyang.ac.kr/handle/20.500.11754/120755 | - |
dc.description.abstract | Series models have several functions: comprehending the functional dependence of variable of interest on covariates, forecasting the dependent variable for future values of covariates and estimating variance disintegration, co-integration and steady-state relations. Although the regression function in a time series model has been extensively modeled both parametrically and nonparametrically, modeling of the error autocorrelation is mainly restricted to the parametric setup. A proper modeling of autocorrelation not only helps to reduce the bias in regression function estimate, but also enriches forecasting via a better forecast of the error term. In this article, we present a nonparametric modeling of autocorrelation function under a Bayesian framework. Moving into the frequency domain from the time domain, we introduce a Gaussian process prior to the log of the spectral density, which is then updated by using a Whittle approximation for the likelihood function (Whittle likelihood). The posterior computation is simplified due to the fact that Whittle likelihood is approximated by the likelihood of a normal mixture distribution with log-spectral density as a location shift parameter, where the mixture is of only five components with known means, variances, and mixture probabilities. The problem then becomes conjugate conditional on the mixture components, and a Gibbs sampler is used to initiate the unknown mixture components as latent variables. We present a simulation study for performance comparison, and apply our method to the two real data examples. | en_US |
dc.description.sponsorship | The authors would like to thank the reviewers and the editors who helped to substantially improve the paper. The authors are indebted to Dr. Nidhan Choudhuri for his continuous encouragement and support. K. H. Kim gratefully acknowledges Hanyang University research fund (HY-2016, HY-2017). C. Lim was supported by Research Resettlement Fund for the new faculty of Seoul National University. C. Lim was also supported by the National Research Foundation of Korea (NRF) Grant funded by the Korea government (MSIP) (No. 2016R1A2B4008237). | en_US |
dc.language.iso | en_US | en_US |
dc.publisher | SPRINGER HEIDELBERG | en_US |
dc.subject | Autocorrelation function | en_US |
dc.subject | Whittle likelihood | en_US |
dc.subject | Bayesian framework | en_US |
dc.subject | Gaussian process prior | en_US |
dc.title | Bayesian time series regression with nonparametric modeling of autocorrelation | en_US |
dc.type | Article | en_US |
dc.relation.no | 4 | - |
dc.relation.volume | 33 | - |
dc.identifier.doi | 10.1007/s00180-018-0796-9 | - |
dc.relation.page | 1715-1731 | - |
dc.relation.journal | COMPUTATIONAL STATISTICS | - |
dc.contributor.googleauthor | Dey, Tanujit | - |
dc.contributor.googleauthor | Kim, Kun Ho | - |
dc.contributor.googleauthor | Lim, Chae Young | - |
dc.relation.code | 2018004816 | - |
dc.sector.campus | S | - |
dc.sector.daehak | COLLEGE OF ECONOMICS AND FINANCE[S] | - |
dc.sector.department | DIVISION OF ECONOMICS & FINANCE | - |
dc.identifier.pid | kunhokim | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.