Comparison of parametric and semiparametric survival regression
models with kernel estimation

J 2021

Comparison of parametric and semiparametric survival regression models with kernel estimation

SELINGEROVÁ, Iveta, Stanislav KATINA a Ivanka HOROVÁ

Základní údaje

Originální název

Comparison of parametric and semiparametric survival regression models with kernel estimation

Autoři

SELINGEROVÁ, Iveta (203 Česká republika, garant, domácí), Stanislav KATINA (703 Slovensko, domácí) a Ivanka HOROVÁ (203 Česká republika, domácí)

Vydání

Journal of Statistical Computation and Simulation, Taylor & Francis, 2021, 0094-9655

Další údaje

Jazyk

angličtina

Typ výsledku

Článek v odborném periodiku

Obor

10103 Statistics and probability

Stát vydavatele

Velká Británie a Severní Irsko

Utajení

není předmětem státního či obchodního tajemství

Odkazy

URL

Impakt faktor

Impact factor: 1.225

Kód RIV

RIV/00216224:14310/21:00121410

Organizační jednotka

Přírodovědecká fakulta

DOI

http://dx.doi.org/10.1080/00949655.2021.1906875

UT WoS

000638231000001

Klíčová slova anglicky

Survival analysis; hazard function; Kernel estimation; simulations; Cox model

Štítky

rivok

Příznaky

Mezinárodní význam, Recenzováno

Změněno: 8. 2. 2022 10:36, Mgr. Tereza Miškechová

Anotace

V originále

The modelling of censored survival data is based on different estimations of the conditional hazard function. When survival time follows a known distribution, parametric models are useful. This strong assumption is replaced by a weaker in the case of semiparametric models. For instance, the frequently used model suggested by Cox is based on the proportionality of hazards. These models use non-parametric methods to estimate some baseline hazard and parametric methods to estimate the influence of a covariate. An alternative approach is to use smoothing that is more flexible. In this paper, two types of kernel smoothing and some bandwidth selection techniques are introduced. Application to real data shows different interpretations for each approach. The extensive simulation study is aimed at comparing different approaches and assessing their benefits. Kernel estimation is demonstrated to be very helpful for verifying assumptions of parametric or semiparametric models and is able to capture changes in the hazard function in both time and covariate directions.

Návaznosti

MUNI/A/1615/2020, interní kód MU

Název: Matematické a statistické modelování 5 (Akronym: MaStaMo5)

Investor: Masarykova univerzita, Matematické a statistické modelování 5

90125, velká výzkumná infrastruktura

Název: BBMRI-CZ III

90128, velká výzkumná infrastruktura

Název: CZECRIN III

Citovat

SELINGEROVÁ, Iveta, Stanislav KATINA a Ivanka HOROVÁ. Comparison of parametric and semiparametric survival regression models with kernel estimation. Journal of Statistical Computation and Simulation. Taylor & Francis, 2021, roč. 91, č. 13, s. 2717-2739. ISSN 0094-9655. Dostupné z: https://dx.doi.org/10.1080/00949655.2021.1906875.

@article{1760377,
   author = {Selingerová, Iveta and Katina, Stanislav and Horová, Ivanka},
   article_number = {13},
   doi = {http://dx.doi.org/10.1080/00949655.2021.1906875},
   keywords = {Survival analysis; hazard function; Kernel estimation; simulations; Cox model},
   language = {eng},
   issn = {0094-9655},
   journal = {Journal of Statistical Computation and Simulation},
   title = {Comparison of parametric and semiparametric survival regression models with kernel estimation},
   url = {https://www.tandfonline.com/doi/full/10.1080/00949655.2021.1906875},
   volume = {91},
   year = {2021}
}

TY  - JOUR
ID  - 1760377
AU  - Selingerová, Iveta - Katina, Stanislav - Horová, Ivanka
PY  - 2021
TI  - Comparison of parametric and semiparametric survival regression models with kernel estimation
JF  - Journal of Statistical Computation and Simulation
VL  - 91
IS  - 13
SP  - 2717-2739
EP  - 2717-2739
PB  - Taylor & Francis
SN  - 00949655
KW  - Survival analysis
KW  - hazard function
KW  - Kernel estimation
KW  - simulations
KW  - Cox model
UR  - https://www.tandfonline.com/doi/full/10.1080/00949655.2021.1906875
N2  - The modelling of censored survival data is based on different estimations of the conditional hazard function. When survival time follows a known distribution, parametric models are useful. This strong assumption is replaced by a weaker in the case of semiparametric models. For instance, the frequently used model suggested by Cox is based on the proportionality of hazards. These models use non-parametric methods to estimate some baseline hazard and parametric methods to estimate the influence of a covariate. An alternative approach is to use smoothing that is more flexible. In this paper, two types of kernel smoothing and some bandwidth selection techniques are introduced. Application to real data shows different interpretations for each approach. The extensive simulation study is aimed at comparing different approaches and assessing their benefits. Kernel estimation is demonstrated to be very helpful for verifying assumptions of parametric or semiparametric models and is able to capture changes in the hazard function in both time and covariate directions.
ER  -

SELINGEROVÁ, Iveta, Stanislav KATINA a Ivanka HOROVÁ. Comparison of parametric and semiparametric survival regression models with kernel estimation. \textit{Journal of Statistical Computation and Simulation}. Taylor \&{} Francis, 2021, roč.~91, č.~13, s.~2717-2739. ISSN~0094-9655. Dostupné z: https://dx.doi.org/10.1080/00949655.2021.1906875.

Podrobný výpis o publikaci