Abstract
The gamma frailty model is a natural extension of the Cox proportional hazards model in survival analysis. Because the frailties are unobserved, an E-M approach is often used for estimation. Such an approach is shown to lead to finite sample underestimation of the frailty variance, with the corresponding regression parameters also being underestimated as a result. For the univariate case, we investigate the source of the bias with simulation studies and a complete enumeration. The rank-based E-M approach, we note, only identifies frailty through the order in which failures occur; additional frailty which is evident in the survival times is ignored, and as a result the frailty variance is underestimated. An adaption of the standard E-M approach is suggested, whereby the non-parametric Breslow estimate is replaced by a local likelihood formulation for the baseline hazard which allows the survival times themselves to enter the model. Simulations demonstrate that this approach substantially reduces the bias, even at small sample sizes. The method developed is applied to survival data from the North West Regional Leukaemia Register.
Article PDF
Similar content being viewed by others
Avoid common mistakes on your manuscript.
References
P. Andersen J. Klein K. Knudsen R. Palacious (1997) ArticleTitle“Estimation of variance in Cox’s regression model with shared gamma frailties” Biometrics 53 1475–1484
R. Betensky L.M. Lindsey J. Ryan M. Wand (2002) ArticleTitle“A local likelihood proportional hazards model for interval censored data” Statistics in Medicine 21 263–275
R. Betensky J. Lindsey L. Ryan M. Wand (1999) ArticleTitle“Local EM estimation of the hazard function for interval-censored data” Biometrics 55 238–245
W. Cleveland (1979) ArticleTitle“Robust locally weighted regression and smoothing scatterplots” Journal of the American Statistical Association 74 829–836
A.P. Dempster N.M. Laird D.B. Rubin (1977) ArticleTitle“Maximum likelihood from incomplete data via the EM algorithm” Journal of the Royal Statistical Society Series B39 1–38
C. Elbers G. Ridder (1982) ArticleTitle“True and spurious duration dependence The identifiability of the proportional hazard model” Review of Economic Studies 49 403–409
J. Heckman B. Singer (1984) ArticleTitleidentifiability of the proportional hazard model” Review of Economic Studies 51 231–241
R. Henderson P. Oman (1999) ArticleTitle“Effect of frailty on marginal regression estimates in survival analysis”. Journal of the Royal Statistical Society Series B 61 367–379
P. Hougaard (2000) Analysis of Multivariate Survival Data Springer-Verlag New York
J. Klein (1992) ArticleTitle“Semiparametric estimation of random effects using the Cox model based on the E-M algorithm” Biometrics 48 795–806
J.P. Klein M.L. Moeschberger (1997) Survival Analysis: Techniques for Censored and Truncated Data Springer-Verlag New York
G.G. Nielsen R.G. Gill P.K. Andersen (1992) ArticleTitle“A counting process approach to maximum likelihood estimation in frailty models” Scandinavian Journal of Statistics 19 25–43
E. Parner (1998) ArticleTitle“Asymptotic theory for the correlated gamma-frailty model” The Annals of Statistics 26 183–214
V. Rondeau D. Commenges P. Joly (2003) ArticleTitle“Maximum penalized lieklihood estimation in a gamma frailty model” Lifetime data analysis 9 139–153
T. Therneau P. Grambsch (2000) Modelling Survival Data Springer-Verlag New York
R. Tibshirani T. Hastie (1987) ArticleTitle“Local likelihood estimation” Journal of the American Statistical Society 82 559–567
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Barker, P., Henderson, R. Small Sample Bias in the Gamma Frailty Model for Univariate Survival. Lifetime Data Anal 11, 265–284 (2005). https://doi.org/10.1007/s10985-004-0387-7
Received:
Revised:
Accepted:
Issue Date:
DOI: https://doi.org/10.1007/s10985-004-0387-7