A nested copula duration model for competing risks with multiple spells

Simon M.S. Lo; Enno Mammen; Ralf A. Wilke

doi:10.1016/j.csda.2020.106986

A nested copula duration model for competing risks with multiple spells

Simon M.S. Lo, Enno Mammen, Ralf A. Wilke

Research output: Contribution to journal › Article › peer-review

13 Citations (Scopus)

Abstract

A copula graphic estimator for the competing risks duration model with multiple spells is presented. By adopting a nested copula structure the dependencies between risks and spells are modelled separately. This breaks up an implicit restriction of popular duration models such as multivariate mixed proportional hazards. It is shown that the dependence structure between spells is identifiable and can be estimated, in contrast to the dependence structure between competing risks. Thus, by allowing these two components to differ, the model is not identifiable. This is an important finding related to the general identifiability of competing risks models. Various features of the model are investigated by simulations and its practicality is illustrated by an application to unemployment duration data.

Original language	English
Article number	106986
Journal	Computational Statistics and Data Analysis
Volume	150
DOIs	https://doi.org/10.1016/j.csda.2020.106986
Publication status	Published - Oct 2020
Externally published	Yes

Keywords

Frailty
Multiple occurrences
Nested archimedean copula

ASJC Scopus subject areas

Statistics and Probability
Computational Mathematics
Computational Theory and Mathematics
Applied Mathematics

UN SDGs

This output contributes to the following UN Sustainable Development Goals (SDGs)

Access to Document

10.1016/j.csda.2020.106986

Cite this

@article{c8d5d94bf07f4b2ba24a1856c427712c,

title = "A nested copula duration model for competing risks with multiple spells",

abstract = "A copula graphic estimator for the competing risks duration model with multiple spells is presented. By adopting a nested copula structure the dependencies between risks and spells are modelled separately. This breaks up an implicit restriction of popular duration models such as multivariate mixed proportional hazards. It is shown that the dependence structure between spells is identifiable and can be estimated, in contrast to the dependence structure between competing risks. Thus, by allowing these two components to differ, the model is not identifiable. This is an important finding related to the general identifiability of competing risks models. Various features of the model are investigated by simulations and its practicality is illustrated by an application to unemployment duration data.",

keywords = "Frailty, Multiple occurrences, Nested archimedean copula",

author = "Lo, {Simon M.S.} and Enno Mammen and Wilke, {Ralf A.}",

note = "Publisher Copyright: {\textcopyright} 2020 Elsevier B.V.",

year = "2020",

month = oct,

doi = "10.1016/j.csda.2020.106986",

language = "English",

volume = "150",

journal = "Computational Statistics and Data Analysis",

issn = "0167-9473",

publisher = "Elsevier B.V.",

}

TY - JOUR

T1 - A nested copula duration model for competing risks with multiple spells

AU - Lo, Simon M.S.

AU - Mammen, Enno

AU - Wilke, Ralf A.

PY - 2020/10

Y1 - 2020/10

N2 - A copula graphic estimator for the competing risks duration model with multiple spells is presented. By adopting a nested copula structure the dependencies between risks and spells are modelled separately. This breaks up an implicit restriction of popular duration models such as multivariate mixed proportional hazards. It is shown that the dependence structure between spells is identifiable and can be estimated, in contrast to the dependence structure between competing risks. Thus, by allowing these two components to differ, the model is not identifiable. This is an important finding related to the general identifiability of competing risks models. Various features of the model are investigated by simulations and its practicality is illustrated by an application to unemployment duration data.

AB - A copula graphic estimator for the competing risks duration model with multiple spells is presented. By adopting a nested copula structure the dependencies between risks and spells are modelled separately. This breaks up an implicit restriction of popular duration models such as multivariate mixed proportional hazards. It is shown that the dependence structure between spells is identifiable and can be estimated, in contrast to the dependence structure between competing risks. Thus, by allowing these two components to differ, the model is not identifiable. This is an important finding related to the general identifiability of competing risks models. Various features of the model are investigated by simulations and its practicality is illustrated by an application to unemployment duration data.

KW - Frailty

KW - Multiple occurrences

KW - Nested archimedean copula

UR - http://www.scopus.com/inward/record.url?scp=85084671494&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85084671494&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2020.106986

DO - 10.1016/j.csda.2020.106986

M3 - Article

AN - SCOPUS:85084671494

SN - 0167-9473

VL - 150

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

M1 - 106986

ER -

A nested copula duration model for competing risks with multiple spells

Abstract

Keywords

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this