Sensitivity Analysis of Copula Survival Models for Clustered Survival Data
2
조회
0
다운로드
1
인용
0
참고문헌
키워드
초록
Copula-based regression models have often been used for the analysis of clustered multivariate survival data. In particular, the Archimedean copula function is useful for modeling such dependency. For likelihood inference, one- and two-stage estimation methods are usually used. In this study, we focused on the behavior of the estimated model parameters against misspecification of Archimedean copular models. For this purpose, we investigated this sensitivity using simulation studies and real data analysis. The simulation results demonstrate that the estimated regression parameters are robust against misspecification of the copula model under the two-stage procedure but that they are not under the one-stage procedure.
Abstract
Copula-based regression models have often been used for the analysis of clustered multivariate survival data. In particular, the Archimedean copula function is useful for modeling such dependency. For likelihood inference, one- and two-stage estimation methods are usually used. In this study, we focused on the behavior of the estimated model parameters against misspecification of Archimedean copular models. For this purpose, we investigated this sensitivity using simulation studies and real data analysis. The simulation results demonstrate that the estimated regression parameters are robust against misspecification of the copula model under the two-stage procedure but that they are not under the one-stage procedure.
참고문헌 (22)
- Hougaard P, "Analysis of multivariate survival data" (2000), Springer
- Ha ID, "Statistical modelling of survival data with random effects: h-likelihood approach" (2017), Springer
- Goethals K, "Frailty models and copulas : similarities and differences", J Appl Stat, 35, pp.1071~1079 (2008)
- Duchateau L, "The frailty model" (2008), Springer
- 홍주영, "Multivariate Bias Correction for Climate Simulation Data, with Application to Precipitation Extremes in Korea", Quantitative Bio-Science, 38(2), pp.121~130 (2019), 자연과학연구소[DOI]
- Lawless JF, "Comparision of semiparametric maximum likelihood estimation and two-stage semiparametric estimation in copula models", Comput Stat Data An, 55, pp.2446~2455 (2011)
- Prenen L, "Extending the Archimedean copula methodology to model multivariate survival data grouped in clusters of variable size", J Roy Stat Soc B, 79, pp.483~505 (2017)
- Prenen L, "Sunclarco:Survival analysis using copulas, R package version 1.0.0", ~ (2017)
- Fleming TR, "Counting processes and survival analysis" (1991), Wiley
- Joe H, "Multivariate models and dependence concepts" (1997), Chapman and Hall
- Gumbel EJ, "Distribution des valeurs extremes en plusieurs dimensions", Publications de l’Institut de Statistique de L’Universite de Paris, 9, pp.171~173 (1960)
- Hofert M, "Likelihood inference for Archimedean copulas in high dimension under known margins", J Multivariate Anal, 110, pp.133~150 (2012)
- Shih JH, "Inferences on the association parameter in copula models for bivariate survival data", Biometrics, 51, pp.1384~1399 (1995)
- Andersen EW, "Two-stage estimation in copula models used in family studies", Lifetime Data Anal, 11, pp.333~350 (2005)
- Wei LJ, "Regression analysis of multivariate incomplete failure time data by modeling marginal distributions", J Am Stat Assoc, 84, pp.1065~1073 (1989)
- Spiekerman CF, "Marginal regression models for multivariate failure time data", J Am Stat Assoc, 93, pp.1164~1175 (1998)
- Lin DY, "Cox regression analysis of multivariate failure time data : The marginal approach", Stat Med, 13, pp.2233~2247 (1994)
- Ha ID, "Profile likelihood approaches for semiparametric copula and frailty models for clustered survival data", J Appl Stat, 46, pp.2553~2571 (2019)
- Marshall AW, "Families of multivariate distributions", J Am Stat Assoc, 83, pp.834~841 (1988)
- Hougaard P, "Survival models for heterogeneous populations derived from stable distributions", Biometrika, 73, pp.387~396 (1986)
- Liu D, "A positive stable frailty model for clustered failure time data with covariate-dependent frailty", Biometrics, 67, pp.8~17 (2011)
- Yau KKW, "ML and REML estimation in survival analysis with time dependent correlated frailty", Stat Med, 17, pp.1201~1213 (1998)