Generalized inverse-Gaussian frailty models with application to TARGET neuroblastoma data

Luiza S.C. Piancastelli; Wagner Barreto-Souza; Vinícius D. Mayrink

doi:10.1007/s10463-020-00774-z

Generalized inverse-Gaussian frailty models with application to TARGET neuroblastoma data

Luiza S.C. Piancastelli, Wagner Barreto-Souza, Vinícius D. Mayrink

King Abdullah University of Science and Technology

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

5 Scopus citations

Abstract

A new class of survival frailty models based on the generalized inverse-Gaussian (GIG) distributions is proposed. We show that the GIG frailty models are flexible and mathematically convenient like the popular gamma frailty model. A piecewise-exponential baseline hazard function is employed, yielding flexibility for the proposed class. Although a closed-form observed log-likelihood function is available, simulation studies show that employing an EM-algorithm is advantageous concerning the direct maximization of this function. Further simulated results address the comparison of different methods for obtaining standard errors of the estimates and confidence intervals for the parameters. Additionally, the finite-sample behavior of the EM-estimators is investigated and the performance of the GIG models under misspecification assessed. We apply our methodology to a TARGET (Therapeutically Applicable Research to Generate Effective Treatments) data about the survival time of patients with neuroblastoma cancer and show some advantages of the GIG frailties over existing models in the literature.

Original language	English (US)
Journal	Annals of the Institute of Statistical Mathematics
DOIs	https://doi.org/10.1007/s10463-020-00774-z
State	Published - Apr 2 2020

Bibliographical note

KAUST Repository Item: Exported on 2020-12-02
Acknowledgements: We thank the Associate Editor and two anonymous Referees for their insightful comments and suggestions that lead to a great improvement of the paper. We also thank the National Cancer Institute (Office of Cancer Genomics) for granting us permission to use the TARGET Neuroblastoma Clinical data for publication. W. Barreto-Souza would also like to acknowledge support for his research from the KAUST Research Fund, NIH 1R01EB028753-01, and the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq-Brazil, Grant number 305543/2018-0). Part of this work is from the Master’s Thesis of Luiza S.C. Piancastelli realized at the Department of Statistics of the Universidade Federal de Minas Gerais, Brazil.

UN SDGs

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

Access to Document

10.1007/s10463-020-00774-z

https://repository.kaust.edu.sa/bitstream/10754/662520/3/GIG%20frailty%20models%20-%20AISM%20-%20Final%20version.pdf

Cite this

@article{b4963f9f5e854f29b4e3c9187530adf5,

title = "Generalized inverse-Gaussian frailty models with application to TARGET neuroblastoma data",

abstract = "A new class of survival frailty models based on the generalized inverse-Gaussian (GIG) distributions is proposed. We show that the GIG frailty models are flexible and mathematically convenient like the popular gamma frailty model. A piecewise-exponential baseline hazard function is employed, yielding flexibility for the proposed class. Although a closed-form observed log-likelihood function is available, simulation studies show that employing an EM-algorithm is advantageous concerning the direct maximization of this function. Further simulated results address the comparison of different methods for obtaining standard errors of the estimates and confidence intervals for the parameters. Additionally, the finite-sample behavior of the EM-estimators is investigated and the performance of the GIG models under misspecification assessed. We apply our methodology to a TARGET (Therapeutically Applicable Research to Generate Effective Treatments) data about the survival time of patients with neuroblastoma cancer and show some advantages of the GIG frailties over existing models in the literature.",

author = "Piancastelli, {Luiza S.C.} and Wagner Barreto-Souza and Mayrink, {Vin{\'i}cius D.}",

note = "KAUST Repository Item: Exported on 2020-12-02 Acknowledgements: We thank the Associate Editor and two anonymous Referees for their insightful comments and suggestions that lead to a great improvement of the paper. We also thank the National Cancer Institute (Office of Cancer Genomics) for granting us permission to use the TARGET Neuroblastoma Clinical data for publication. W. Barreto-Souza would also like to acknowledge support for his research from the KAUST Research Fund, NIH 1R01EB028753-01, and the Conselho Nacional de Desenvolvimento Cient{\'i}fico e Tecnol{\'o}gico (CNPq-Brazil, Grant number 305543/2018-0). Part of this work is from the Master{\textquoteright}s Thesis of Luiza S.C. Piancastelli realized at the Department of Statistics of the Universidade Federal de Minas Gerais, Brazil.",

year = "2020",

month = apr,

day = "2",

doi = "10.1007/s10463-020-00774-z",

language = "English (US)",

journal = "Annals of the Institute of Statistical Mathematics",

issn = "1572-9052",

publisher = "Springer Netherlands",

}

TY - JOUR

T1 - Generalized inverse-Gaussian frailty models with application to TARGET neuroblastoma data

AU - Piancastelli, Luiza S.C.

AU - Barreto-Souza, Wagner

AU - Mayrink, Vinícius D.

N1 - KAUST Repository Item: Exported on 2020-12-02 Acknowledgements: We thank the Associate Editor and two anonymous Referees for their insightful comments and suggestions that lead to a great improvement of the paper. We also thank the National Cancer Institute (Office of Cancer Genomics) for granting us permission to use the TARGET Neuroblastoma Clinical data for publication. W. Barreto-Souza would also like to acknowledge support for his research from the KAUST Research Fund, NIH 1R01EB028753-01, and the Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq-Brazil, Grant number 305543/2018-0). Part of this work is from the Master’s Thesis of Luiza S.C. Piancastelli realized at the Department of Statistics of the Universidade Federal de Minas Gerais, Brazil.

PY - 2020/4/2

Y1 - 2020/4/2

N2 - A new class of survival frailty models based on the generalized inverse-Gaussian (GIG) distributions is proposed. We show that the GIG frailty models are flexible and mathematically convenient like the popular gamma frailty model. A piecewise-exponential baseline hazard function is employed, yielding flexibility for the proposed class. Although a closed-form observed log-likelihood function is available, simulation studies show that employing an EM-algorithm is advantageous concerning the direct maximization of this function. Further simulated results address the comparison of different methods for obtaining standard errors of the estimates and confidence intervals for the parameters. Additionally, the finite-sample behavior of the EM-estimators is investigated and the performance of the GIG models under misspecification assessed. We apply our methodology to a TARGET (Therapeutically Applicable Research to Generate Effective Treatments) data about the survival time of patients with neuroblastoma cancer and show some advantages of the GIG frailties over existing models in the literature.

AB - A new class of survival frailty models based on the generalized inverse-Gaussian (GIG) distributions is proposed. We show that the GIG frailty models are flexible and mathematically convenient like the popular gamma frailty model. A piecewise-exponential baseline hazard function is employed, yielding flexibility for the proposed class. Although a closed-form observed log-likelihood function is available, simulation studies show that employing an EM-algorithm is advantageous concerning the direct maximization of this function. Further simulated results address the comparison of different methods for obtaining standard errors of the estimates and confidence intervals for the parameters. Additionally, the finite-sample behavior of the EM-estimators is investigated and the performance of the GIG models under misspecification assessed. We apply our methodology to a TARGET (Therapeutically Applicable Research to Generate Effective Treatments) data about the survival time of patients with neuroblastoma cancer and show some advantages of the GIG frailties over existing models in the literature.

UR - http://hdl.handle.net/10754/662520

UR - http://link.springer.com/10.1007/s10463-020-00774-z

UR - http://www.scopus.com/inward/record.url?scp=85096468398&partnerID=8YFLogxK

U2 - 10.1007/s10463-020-00774-z

DO - 10.1007/s10463-020-00774-z

M3 - Article

SN - 1572-9052

JO - Annals of the Institute of Statistical Mathematics

JF - Annals of the Institute of Statistical Mathematics

ER -

Generalized inverse-Gaussian frailty models with application to TARGET neuroblastoma data

Abstract

Bibliographical note

UN SDGs

Access to Document

Other files and links

Fingerprint

Cite this