Lifetime Data Analysis

, Volume 8, Issue 2, pp 163–176 | Cite as

Hierarchical-Likelihood Approach for Mixed Linear Models with Censored Data

  • Il Do Ha
  • Youngjo Lee
  • Jae-Kee Song


Mixed linear models describe the dependence via random effects in multivariate normal survival data. Recently they have received considerable attention in the biomedical literature. They model the conditional survival times, whereas the alternative frailty model uses the conditional hazard rate. We develop an inferential method for the mixed linear model via Lee and Nelder's (1996) hierarchical-likelihood (h-likelihood). Simulation and a practical example are presented to illustrate the new method.

hierarchical-likelihood mixed linear model multivariate survival data random effects restricted maximum likelihood 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. M. A. Aitkin, “A note on the regression analysis of censored data,” Technometrics vol. 23 pp. 161–163, 1981.Google Scholar
  2. J. Buckley and I. James, “Linear regression with censored data,” Biometrika vol. 66 pp. 429–436, 1979.Google Scholar
  3. S. M. Butler and T. A. Louis, “Random effects models with non-parametric priors,” Statistics in Medicine vol. 11 pp. 1981–2000, 1992.Google Scholar
  4. D. R. Cox, “Regression models and life tables (with discussion),” Journal of the Royal Statistical Society B vol. 74 pp. 187–220, 1972.Google Scholar
  5. I. D. Ha, Y. Lee, and J.-K. Song, “Hierarchical likelihood approach for frailty models,” Biometrika vol. 88 pp. 233–243, 2001.Google Scholar
  6. C. R. Henderson, “Best linear unbiased estimation and prediction under a selection model,” Biometrics vol. 31 pp. 423–447, 1975.Google Scholar
  7. J. P. Hughes, “Mixed effects models with censored data with application to HIV RNA levels,” Biometrics vol. 55 pp. 625–629, 1999.Google Scholar
  8. J. P. Klein and M. L. Moeschberger, Survival Analysis: Techniques for Censored and Truncated Data, New York: Springer-Verlag, 1997.Google Scholar
  9. J. P. Klein, C. Pelz, and M. Zhang, “Modelling random effects for censored data by a multivariate normal regression model,” Biometrics vol. 55 pp. 497–506, 1999.Google Scholar
  10. Y. Lee and J. A. Nelder, “Hierarchical generalized linear models (with discussion),” Journal of the Royal Statistical Society B vol. 58 pp. 619–678, 1996.Google Scholar
  11. Y. Lee and J. A. Nelder, “Hierarchical generalized linear models: A synthesis of generalized linear models, random-effect models and structured dispersions,” Technical report, Department of Mathematics, Imperial College, London, 2000.Google Scholar
  12. N. Mantel, N. R. Bohidar, and J. L. Ciminera, “Mantel-Haenszel analysis of litter-matched time-to-response data, with modifications for recovery of interlitter information,” Cancer Research vol. 37 pp. 3863–3868, 1977.Google Scholar
  13. C. A. McGilchrist, “REML estimation for survival models with frailty,” Biometrics vol. 49 pp. 221- 225, 1993.Google Scholar
  14. G. G. Nielsen, R. D. Gill, P. K. Andersen, and T. I. A. Sørensen, “A counting process approach to maximum likelihood estimation in frailty models,” Scandinavian Journal of Statistics vol. 19 pp. 25–44, 1992.Google Scholar
  15. W. Pan and C. Kooperberg, “Linear regression for bivariate censored data via multiple imputation,” Statistics in Medicine vol. 18 pp. 3111–3121, 1999.Google Scholar
  16. W. Pan and T. A. Louis, “A linear mixed-effects model for multivariate censored data,” Biometrics vol. 56 pp. 160–166, 2000.Google Scholar
  17. H. D. Patterson and R. Thompson, “Recovery of interblock information when block sizes are unequal,” Biometrika vol. 58 pp. 545–554, 1971.Google Scholar
  18. A. N. Pettitt, “Censored observations, repeated measures and mixed effects models: An approach using the EM algorithm and normal errors,” Biometrika vol. 73 pp. 635–643, 1986.Google Scholar
  19. J. Schmee and G. J. Hahn, “A simple method for regression analysis with censored data,” Technometrics vol. 21 pp. 417–423, 1979.Google Scholar
  20. G. Verbeke and E. Lesaffre, “The effects of misspecifying the random-effects distribution in linear mixed models for longitudinal data,” Computational Statistics and Data Analysis vol. 23 pp. 541–556, 1997.Google Scholar
  21. M. S. Wolynetz, “Maximum likelihood estimation in a linear model from confined and censored normal data (Algorithm AS139),” Applied Statistics vol. 28 pp. 195–206, 1979.Google Scholar

Copyright information

© Kluwer Academic Publishers 2002

Authors and Affiliations

  • Il Do Ha
    • 1
  • Youngjo Lee
    • 2
  • Jae-Kee Song
    • 3
  1. 1.Department of StatisticsKyungsan UniversityKyungsanSouth Korea
  2. 2.Department of StatisticsSeoul National UniversitySeoulSouth Korea
  3. 3.Department of StatisticsKyungpook National UniversityTaeguSouth Korea

Personalised recommendations