A Nested Copula Duration Model for Competing Risks with Multiple Spells

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

Department of Economics

Research output: Contribution to journal › Journal article › Research › peer-review

111 Downloads (Pure)

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 & Data Analysis
Volume	150
Number of pages	32
ISSN	0167-9473
DOIs	https://doi.org/10.1016/j.csda.2020.106986
Publication status	Published - Oct 2020

Bibliographical note

Published online: 25 April 2020

Keywords

Nested archimedean coupla
Multiple occurrences
Frailty

UN SDGs

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

Access to Document

10.1016/j.csda.2020.106986Licence: Unspecified

ralf_wilke_et_al_a_nested_copula_acceptedmanuscriptAccepted author manuscript, 513 KBLicence: CC BY-NC-ND

Handle.net

Persistent link

Cite this

@article{5c8572d8168640048d91eb5aa6debe23,

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 = "Nested archimedean coupla, Multiple occurrences, Frailty, Nested archimedean coupla, Multiple occurrences, Frailty",

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

note = "Published online: 25 April 2020",

year = "2020",

month = oct,

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

language = "English",

volume = "150",

journal = "Computational Statistics & Data Analysis",

issn = "0167-9473",

publisher = "Elsevier",

}

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

N1 - Published online: 25 April 2020

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 - Nested archimedean coupla

KW - Multiple occurrences

KW - Frailty

KW - Nested archimedean coupla

KW - Multiple occurrences

KW - Frailty

U2 - 10.1016/j.csda.2020.106986

DO - 10.1016/j.csda.2020.106986

M3 - Journal article

SN - 0167-9473

VL - 150

JO - Computational Statistics & Data Analysis

JF - Computational Statistics & Data Analysis

M1 - 106986

ER -