Skip to main content

Generalized linear mixed models with informative dropouts and missing covariates

Abstract

Generalized linear mixed models (GLMM) are useful in many longitudinal data analyses. In the presence of informative dropouts and missing covariates, however, standard complete-data methods may not be applicable. In this article, we consider a likelihood method and an approximate method for GLMM with informative dropouts and missing covariates. The methods are implemented by Monte–Carlo EM algorithms combined with Gibbs sampler. The approximate method may lead to inconsistent estimators but is computationally more efficient than the likelihood method. The two methods are evaluated via a simulation study for longitudinal binary data, and appear to perform reasonably well. A dataset on mental distress is analyzed in details.

This is a preview of subscription content, access via your institution.

References

  1. Booth JG, Hobert JP (1999) Maximizing generalized linear mixed model likelihoods with an automated Monte Carlo EM algorithm. J R Stat Soc Ser B 61:265–285

    MATH  Article  Google Scholar 

  2. Breslow NE, Clayton DG (1993) Approximate inference in generalized linear mixed models. J Am Stat Assoc 88:9–25

    MATH  Article  Google Scholar 

  3. Caffo BS, Jank W, Jones GL (2005) Ascent-based Monte Carlo expectation - maximization. J R Stat Soc Ser B 67:235–251

    MATH  Article  MathSciNet  Google Scholar 

  4. Chen MH, Ibrahim JG, Shao QM (2004) Propriety of the posterior distribution and existence of the maximum likelihood estimator for regression models with covariates missing at random. J Am Stat Assoc 99:421–438

    Article  MathSciNet  Google Scholar 

  5. Diggle PJ, Heagerty P, Liang KY, Zeger SL (2002) Analysis of longitudinal data, 2nd edn. Oxford University Press, New York

    Google Scholar 

  6. van Dyk DA (2000) Fitting mixed-effects models using efficient EM-type algorithms. J Comput Graph Stat 9:78–98

    Article  Google Scholar 

  7. Fort G, Moulines E (2003) Convergence of the Monte-Carlo EM for curved exponential families. Ann Stat 31(4):1220–1259

    MATH  Article  MathSciNet  Google Scholar 

  8. Gelfand AE, Sahu SK, Carlin BP (1996) Efficient parametrisations for generalized linear mixed models (with discussion). In: Bernardo JM, Berger JO, Dawid AP, Smith AFM (eds) Bayesian statistics, vol 5. Oxford University Press, New York, pp 165–180

    Google Scholar 

  9. Geweke J (1996) Handbook of computational economics, Chap 15. North-Holland, Amsterdam

    Google Scholar 

  10. Gilks WR, Wild P (1992) Adaptive rejection sampling for Gibbs sampling. Appl Stat 41:337–348

    MATH  Article  Google Scholar 

  11. Ibrahim JG, Lipsitz SR, Chen MH (1999) Missing covariates in generalized linear models when the missing data mechanism is nonignorable. J R Stat Soc Ser B 61:173–190

    MATH  Article  MathSciNet  Google Scholar 

  12. Ibrahim JG, Chen MH, Lipsitz SR (2001) Missing responses in generalized linear mixed models when the missing data mechanism is nonignorable. Biometrika 88:551–564

    MATH  Article  MathSciNet  Google Scholar 

  13. Jiang J (1998) Consistent estimators in generalized linear mixed models. J Am Stat Assoc 93: 720–729

    MATH  Article  Google Scholar 

  14. Laird NM, Ware JH (1982) Random-effects models for longitudinal data. Biometrics 38:963–974

    MATH  Article  Google Scholar 

  15. Levine RA, Casella G (2001) Implementations of the Monte Carlo EM algorithm. J Comput Graph Stat 10:422–439

    Article  MathSciNet  Google Scholar 

  16. Little RJA (1995) Modeling the drop-out mechanism in repeated-measures studies. J Am Stat Assoc 90:1112–1121

    MATH  Article  MathSciNet  Google Scholar 

  17. Liu C, Rubin DB, Wu YN (1998) Parameter expansion for EM acceleration: the PX-EM algorithm. Biometrika 85:755–770

    MATH  Article  MathSciNet  Google Scholar 

  18. Louis TA (1982) Finding the observed information matrix when using the EM algorithm. J R Stat Soc Ser B 44:226–233

    MATH  MathSciNet  Google Scholar 

  19. McCulloch CE (1997) Maximum likelihood algorithms for generalized linear mixed models. J Am Stat Assoc 92:162–170

    MATH  Article  MathSciNet  Google Scholar 

  20. McCulloch CE, Searle SR (2001) Generalized, linear, and mixed models. Wiley, New York

    MATH  Google Scholar 

  21. McLachlan GJ, Krishnan T (1997) The EM-algorithm and extensions. Wiley, New York

    MATH  Google Scholar 

  22. Meng XL, van Dyk DA (1997) The EM algorithm – an old folk-song sung to a fast new tune (with discussion). J R Stat Soc Ser B 59:511–567

    MATH  Article  Google Scholar 

  23. Murphy S, Das Gupta A, Cain KC, Johnson LC, Lohan J, Wu L, Mekwa J (1999) Changes in parents’ mental distress after the violent death of an adolescent or young adult child: a longitudinal prospective analysis. Death Stud 23:129–159

    Article  Google Scholar 

  24. Stubbendick AL, Ibrahim JG (2003) Maximum likelihood methods for nonignorable missing responses and covariates in random effects models. Biometrics 59:1140–1150

    MATH  Article  MathSciNet  Google Scholar 

  25. Ten Have TR, Pulkstenis E, Kunselman A, Landis JR (1998) Mixed effects logistics regression models for longitudinal binary response data with informative dropout. Biometrics 54:367–383

    MATH  Article  Google Scholar 

  26. Vonesh EF, Wang H, Nie L, Majumdar D (2002) Conditional second-order generalized estimating equations for generalized linear and nonlinear mixed-effects models. J Am Stat Assoc 97:271–283

    MATH  Article  MathSciNet  Google Scholar 

  27. Wei GC, Tanner MA (1990) A Monte Carlo implementation of the EM algorithm and the poor man’s data augmentation algorithm. J Am Stat Assoc 85:699–704

    Article  Google Scholar 

  28. Wu L (2004) Exact and approximate inferences for nonlinear mixed-effects models with missing covariates. J Am Stat Assoc 99:700–709

    Article  Google Scholar 

Download references

Author information

Affiliations

Authors

Corresponding author

Correspondence to Lang Wu.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Wu, K., Wu, L. Generalized linear mixed models with informative dropouts and missing covariates. Metrika 66, 1–18 (2007). https://doi.org/10.1007/s00184-006-0083-6

Download citation

Keywords

  • PX-EM algorithm
  • Gibbs sampling
  • Linearization
  • Rejection sampling