Statistics and Computing

, Volume 19, Issue 2, pp 129–138 | Cite as

A new method for the estimation of variance matrix with prescribed zeros in nonlinear mixed effects models

  • Djalil ChafaïEmail author
  • Didier Concordet


We propose a new method for the Maximum Likelihood Estimator (MLE) of nonlinear mixed effects models when the variance matrix of Gaussian random effects has a prescribed pattern of zeros (PPZ). The method consists of coupling the recently developed Iterative Conditional Fitting (ICF) algorithm with the Expectation Maximization (EM) algorithm. It provides positive definite estimates for any sample size, and does not rely on any structural assumption concerning the PPZ. It can be easily adapted to many versions of EM.


Nonlinear mixed effects models Maximum likelihood Expected maximisation algorithm Longitudinal data analysis Repeated measurements Iterated proportional fitting algorithm Gaussian graphical models Stochastic inverse problems Pharmacokinetic/pharmacodynamics analysis 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Anderson, T.W.: Asymptotically efficient estimation of covariance matrices with linear structure. Ann. Stat. 1, 135–141 (1973) zbMATHCrossRefGoogle Scholar
  2. Beal, S.L., Sheiner, L.B.: NONMEM user’s guide. Nonlinear mixed effects models for repeated measures data. University of California, San Francisco (1992) Google Scholar
  3. Cepeda, E.C., Gamerman, D.: Bayesian modeling of joint regressions for the mean and covariance matrix. Biom. J. 46(4), 430–440 (2004) CrossRefMathSciNetGoogle Scholar
  4. Chaudhuri, S., Drton, M., Richardson, T.S.: Estimation of a covariance matrix with zeros. Biometrika 94(1), 199–216 (2007) zbMATHCrossRefMathSciNetGoogle Scholar
  5. Concordet, D., Nunez, O.G.: A simulated pseudo-maximum likelihood estimator for nonlinear mixed models. Comput. Stat. Data Anal. 39(2), 187–201 (2002) zbMATHCrossRefMathSciNetGoogle Scholar
  6. Cox, D.R., Wermuth, N., Marchetti, G.: Decompositions and estimation of a chain of covariances. Technical Report, Department of Mathematical Statistics, Chalmers Goteborgs Universitet (2004) Google Scholar
  7. Dahl, J., Vandenberghe, L., Roychowdhury, V.: Covariance selection for non-chordal graphs via chordal embedding. Preprint, available on (2006)
  8. Davidian, M., Giltinan, D.M.: Nonlinear Models for Repeated Measurement Data. Chapman and Hall, New York (1995) Google Scholar
  9. Dempster, A.P., Laird, N.M., Rubin, D.B.: Maximum likelihood from incomplete data via the EM algorithm. J. R. Stat. Soc. Ser. B 39(1), 1–38 (1977). With discussion zbMATHMathSciNetGoogle Scholar
  10. Drton, M., Perlman, M.D.: Model selection for Gaussian concentration graphs. Biometrika 91, 591–602 (2004) zbMATHCrossRefMathSciNetGoogle Scholar
  11. Horn, R.A., Johnson, C.R.: Matrix Analysis. Cambridge University Press, Cambridge (1990). Corrected reprint of the 1985 original zbMATHGoogle Scholar
  12. Jamshidian, M., Jennrich, R.I.: Standard errors for em estimation. J. R. Stat. Soc. B 62, 257–270 (2000) CrossRefMathSciNetGoogle Scholar
  13. Kuhn, E., Lavielle, M.: Maximum likelihood estimation in nonlinear mixed effects models. Comput. Stat. Data Anal. 49(4), 1020–1038 (2005) zbMATHCrossRefMathSciNetGoogle Scholar
  14. Lindstrom, M.J., Bates, D.M.: Nonlinear mixed effects models for repeated measures data. Biometrics 46, 673–687 (1990) CrossRefMathSciNetGoogle Scholar
  15. Pinheiro, J.C., Bates, D.M.: Approximation of the log-likelihood function in the nonlinear mixed effects models. J. Comput. Graph. Stat. 4, 12–35 (1995) CrossRefGoogle Scholar
  16. Pourahmadi, M.: Joint mean-covariance models with applications to longitudinal data: unconstrained parameterisation. Biometrika 86(3), 677–690 (1999) zbMATHCrossRefMathSciNetGoogle Scholar
  17. Pourahmadi, M.: Maximum likelihood estimation of generalised linear models for multivariate normal covariance matrix. Biometrika 87(2), 425–435 (2000) zbMATHCrossRefMathSciNetGoogle Scholar
  18. Vonesh, E.F., Carter, R.L.: Mixed-effects nonlinear regression for unbalanced repeated measures. Biometrics 48, 1–17 (1992) CrossRefMathSciNetGoogle Scholar
  19. Wang, J.: EM algorithms for nonlinear mixed effects models. Comput. Stat. Data Anal. 51(6), 3244–3256 (2007) CrossRefGoogle Scholar
  20. Wolfinger, R.D.: Laplace’s approximation for nonlinear mixed models. Biometrika 80(4), 791–795 (1993) zbMATHCrossRefMathSciNetGoogle Scholar
  21. Wolfinger, R.D.: Heterogeneous variance-covariance structures for repeated measures. J. Agric. Biol. Environ. Stat. 1(2), 205–230 (1996) CrossRefMathSciNetGoogle Scholar
  22. Wu, C.F.: On the convergence properties of the EM algorithm. Ann. Stat. 11, 95–103 (1983) zbMATHCrossRefGoogle Scholar
  23. Ye, H., Pan, J.: Modelling of covariance structures in generalised estimating equations for longitudinal data. Biometrika 93, 927–941 (2006) CrossRefMathSciNetGoogle Scholar
  24. Zhang, F. (ed.): The Schur Complement and its Applications. Numerical Methods and Algorithms, vol. 4. Springer, New York (2005) zbMATHGoogle Scholar
  25. Zimmerman, D.L., Núñez-Antón, V.: Parametric modelling of growth curve data: an overview. Test 10(1), 1–73 (2001). With comments and a rejoinder by the authors zbMATHCrossRefMathSciNetGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2008

Authors and Affiliations

  1. 1.UMR181, INRA, ENVTToulouse CEDEX 3France
  2. 2.UMR CNRS 5219, Institut de Mathématiques de ToulouseUniversité Paul SabatierToulouse CEDEX 9France

Personalised recommendations