File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1016/j.ress.2017.05.029
- Scopus: eid_2-s2.0-85019845810
- WOS: WOS:000412607200011
- Find via
Supplementary
- Citations:
- Appears in Collections:
Article: A nonparametric Bayesian modeling approach for heterogeneous lifetime data with covariates
Title | A nonparametric Bayesian modeling approach for heterogeneous lifetime data with covariates |
---|---|
Authors | |
Keywords | Bayesian modeling Gibbs sampler Lifetime regression Mixture model Model selection |
Issue Date | 2017 |
Citation | Reliability Engineering and System Safety, 2017, v. 167, p. 95-104 How to Cite? |
Abstract | Lifetime data collected at product design and production stage or field operational stage often exhibit heterogeneity patterns, making the homogeneity assumption in conventional statistical lifetime models invalid. Mixture models are important modeling approaches that account for data heterogeneity. However, existing mixture models are constrained by assuming an known number of sub-populations. This paper proposes a new Bayesian statistical model to analyze heterogeneous lifetime data by assuming an unknown number of sub-populations. Each sub-population is characterized by an accelerated failure time model to quantify the effects of possible reliability impact factors. The proposed model allows simultaneous identification of the number of sub-populations and the model parameters of sub-populations. Convenient sampling strategies are further proposed to address the challenges of model estimation. Both numerical case study and real case study are provided to illustrate the proposed approach and demonstrate its validity. |
Persistent Identifier | http://hdl.handle.net/10722/330550 |
ISSN | 2023 Impact Factor: 9.4 2023 SCImago Journal Rankings: 2.028 |
ISI Accession Number ID |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Li, Mingyang | - |
dc.contributor.author | Meng, Hongdao | - |
dc.contributor.author | Zhang, Qingpeng | - |
dc.date.accessioned | 2023-09-05T12:11:43Z | - |
dc.date.available | 2023-09-05T12:11:43Z | - |
dc.date.issued | 2017 | - |
dc.identifier.citation | Reliability Engineering and System Safety, 2017, v. 167, p. 95-104 | - |
dc.identifier.issn | 0951-8320 | - |
dc.identifier.uri | http://hdl.handle.net/10722/330550 | - |
dc.description.abstract | Lifetime data collected at product design and production stage or field operational stage often exhibit heterogeneity patterns, making the homogeneity assumption in conventional statistical lifetime models invalid. Mixture models are important modeling approaches that account for data heterogeneity. However, existing mixture models are constrained by assuming an known number of sub-populations. This paper proposes a new Bayesian statistical model to analyze heterogeneous lifetime data by assuming an unknown number of sub-populations. Each sub-population is characterized by an accelerated failure time model to quantify the effects of possible reliability impact factors. The proposed model allows simultaneous identification of the number of sub-populations and the model parameters of sub-populations. Convenient sampling strategies are further proposed to address the challenges of model estimation. Both numerical case study and real case study are provided to illustrate the proposed approach and demonstrate its validity. | - |
dc.language | eng | - |
dc.relation.ispartof | Reliability Engineering and System Safety | - |
dc.subject | Bayesian modeling | - |
dc.subject | Gibbs sampler | - |
dc.subject | Lifetime regression | - |
dc.subject | Mixture model | - |
dc.subject | Model selection | - |
dc.title | A nonparametric Bayesian modeling approach for heterogeneous lifetime data with covariates | - |
dc.type | Article | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1016/j.ress.2017.05.029 | - |
dc.identifier.scopus | eid_2-s2.0-85019845810 | - |
dc.identifier.volume | 167 | - |
dc.identifier.spage | 95 | - |
dc.identifier.epage | 104 | - |
dc.identifier.isi | WOS:000412607200011 | - |