, Volume 75, Issue 7, pp 913–938 | Cite as

Robust analysis of longitudinal data with nonignorable missing responses

  • Sanjoy K. SinhaEmail author


We encounter missing data in many longitudinal studies. When the missing data are nonignorable, it is important to analyze the data by incorporating the missing data mechanism into the observed data likelihood function. The classical maximum likelihood (ML) method for analyzing longitudinal missing data has been extensively studied in the literature. However, it is well-known that the ordinary ML estimators are sensitive to extreme observations or outliers in the data. In this paper, we propose and explore a robust method, which is developed in the framework of the ML method, and is useful for downweighting any influential observations in the data when estimating the model parameters. We study the empirical properties of the robust estimators in small simulations. We also illustrate the robust method using incomplete longitudinal data on CD4 counts from clinical trials of HIV-infected patients.


Generalized linear models Incomplete data Missing responses Mixed models Robust estimation 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Baker SG, Laird NM (1988) Regression analysis for categorical variables with outcome subject to nonresponse. J Am Stat Assoc 83: 62–69MathSciNetCrossRefGoogle Scholar
  2. Beaumont JF (1999) A robust estimation method in the presence of nonignorable nonresponse. In: Proceedings of the section on survey research methods. American Statistical Association, pp 819–824Google Scholar
  3. Brown CH (1990) Protecting against nonrandomly missing data in longitudinal studies. Biometrics 46: 143–157zbMATHCrossRefGoogle Scholar
  4. Cantoni E, Ronchetti E (2001) Robust inference for generalized linear models. J Am Stat Assoc 96: 1022–1030MathSciNetzbMATHCrossRefGoogle Scholar
  5. Dantan E, Proust-Lima C, Letenneur L, Jacqmin-Gadda H (2008) Pattern mixture models and latent class models for the analysis of multivariate longitudinal data with informative dropouts. Int J Biostat 4 (article 14)Google Scholar
  6. Diggle P, Kenward MG (1994) Informative dropout in longitudinal data analysi (with discussion). Appl Stat 43: 49–94zbMATHCrossRefGoogle Scholar
  7. Gallant JE, Moore RD, Richman DD, Keruly J, Chaisson RE (1992) Incidence and natural history of cytomegalovirus disease in patients with advanced human immunodeficiency virus disease treated with Zidovudine. J Inf Dis 166: 1223–1227CrossRefGoogle Scholar
  8. Ibrahim JG, Lipsitz SR, Chen MH (1999) Missing covariates in generalized linear models when the missing data mechanism is non-ignorable. J R Stat Soc Ser B 61: 173–190MathSciNetzbMATHCrossRefGoogle Scholar
  9. Ibrahim JG, Chen MH, Lipsitz SR (2001) Missing responses in generalized linear mixed models when the missing data mechanism is nonignorable. Biometrika 88: 551–564MathSciNetzbMATHCrossRefGoogle Scholar
  10. Kahn JO, Lagakos SW, Richman DD (1992) A controlled trial comparing continued zidovudine with didanosine in human immunodeficiency virus infection. New Eng J Med 327: 581–587CrossRefGoogle Scholar
  11. Little RJA (1988) Robust estimation of the mean and covariance matrix from data with missing values. Appl Stat 37: 23–38MathSciNetzbMATHCrossRefGoogle Scholar
  12. Little RJA (1995) Modeling the drop-Out mechanism in repeated-measures studies. J Am Stat Assoc 90: 1112–1121MathSciNetzbMATHCrossRefGoogle Scholar
  13. Little RJA, Rubin DB (2002) Statistical Analysis with missing data, 2nd edn. Wiley, New JerseyzbMATHGoogle Scholar
  14. McCulloch CE (1997) Maximum likelihood algorithms for generalized linear mixed models. J Am Stat Assoc 92: 162–170MathSciNetzbMATHCrossRefGoogle Scholar
  15. Molenberghs G, Verbeke G (2001) A review on linear mixed models for longitudinal data, possibly subject to dropout. Stat Modell 1: 235–269zbMATHCrossRefGoogle Scholar
  16. Preisser JS, Galecki AT, Lohman KK, Wagenknecht LE (2000) Analysis of smoking trends with incomplete longitudinal binary responses. J Am Stat Assoc 95: 1021–1031MathSciNetzbMATHCrossRefGoogle Scholar
  17. Preisser JS, Qaqish BF (1999) Robust regression to clustered data with application to binary responses. Biometrics 55: 574–579zbMATHCrossRefGoogle Scholar
  18. Rousseeuw PJ, van Zomeren BC (1990) Unmasking multivariate outliers and leverage points. J Am Stat Assoc 85: 633–639CrossRefGoogle Scholar
  19. Rubin DB (1976) Inference and missing data. Biometrika 63: 581–592MathSciNetzbMATHCrossRefGoogle Scholar
  20. Sinha SK (2004) Robust analysis of generalized linear mixed models. J Am Stat Assoc 99: 451–460zbMATHCrossRefGoogle Scholar
  21. Sinha SK (2008) Robust methods for generalized linear models with nonignorable missing covariates. Can J Stat 36(2): 277–299zbMATHCrossRefGoogle Scholar
  22. Verbeke G, Molenberghs G (2005) Longitudinal and incomplete clinical studies. Metron 63: 143–170MathSciNetGoogle Scholar
  23. Wu L, Liu W, Liu J (2009) A longitudinal study of children’s aggressive behaviours based on multivariate mixed models with incomplete data. Can J Stat 37: 435–452zbMATHCrossRefGoogle Scholar
  24. Xie H (2008) A local sensitivity analysis approach to longitudinal non-gaussian data with non-ignorable dropout. Stat Med 27: 3155–3177MathSciNetCrossRefGoogle Scholar
  25. Yi GY, Cook RJ (2002) Marginal methods for incomplete longitudinal data arising in clusters. J Am Stat Assoc 97: 1071–1080MathSciNetzbMATHCrossRefGoogle Scholar

Copyright information

© Springer-Verlag 2011

Authors and Affiliations

  1. 1.School of Mathematics and StatisticsCarleton UniversityOttawaCanada

Personalised recommendations