Full metadata record
DC Field | Value | Language |
---|---|---|
dc.contributor.author | 김성욱 | - |
dc.date.accessioned | 2019-10-23T01:47:53Z | - |
dc.date.available | 2019-10-23T01:47:53Z | - |
dc.date.issued | 2005-08 | - |
dc.identifier.citation | CSAM(Communications for Statistical Applications and Methods), v. 12, No. 2, Page. 483-496 | en_US |
dc.identifier.issn | 2287-7843 | - |
dc.identifier.uri | http://kiss.kstudy.com/thesis/thesis-view.asp?key=2462173 | - |
dc.identifier.uri | https://repository.hanyang.ac.kr/handle/20.500.11754/111403 | - |
dc.description.abstract | We consider the power law process which is assumed to have multiple changepoints. We propose a binary segmentation procedure for locating all existing changepoints. We select one model between the no-changepoints model and the single changepoint model by the Bayes factor. We repeat this procedure until no more changepoints are found. Then we carry out a multiple test based on the Bayes factor through the intrinsic priors of Berger and Pericchi (1996) to investigate the system behaviour of failure times. We demonstrate our procedure with a real dataset and some simulated datasets. | en_US |
dc.language.iso | en_US | en_US |
dc.publisher | 한국통계학회 | en_US |
dc.subject | Binary segmentation | en_US |
dc.subject | Changepoint | en_US |
dc.subject | Model selection | en_US |
dc.subject | Intrinsic prior | en_US |
dc.subject | Power Law process | en_US |
dc.title | Bayesian changepoints detection for the power law process with binary segmentation procedures | en_US |
dc.type | Article | en_US |
dc.relation.journal | 한국통계학회 논문집 | - |
dc.contributor.googleauthor | Kim, Hyun Soo | - |
dc.contributor.googleauthor | Kim, Seong W. | - |
dc.contributor.googleauthor | Jang, Hak Jin | - |
dc.relation.code | 2012211008 | - |
dc.sector.campus | E | - |
dc.sector.daehak | COLLEGE OF SCIENCE AND CONVERGENCE TECHNOLOGY[E] | - |
dc.sector.department | DEPARTMENT OF APPLIED MATHEMATICS | - |
dc.identifier.pid | seong | - |
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.