Robust inference for mixed censored and binary response models with missing covariates

Abstract

In biomedical and epidemiological studies, often outcomes obtained are of mixed discrete and continuous in nature. Furthermore, due to some technical inconvenience or else, continuous responses are censored and also a few covariates cease to be observed completely. In this paper, we develop a model to tackle these complex situations. Our methodology is developed in a more general framework and provides a full-scale robust analysis of such complex models. The proposed robust maximum likelihood estimators of the model parameters are resistant to potential outliers in the data. We discuss the asymptotic properties of the robust estimators. To avoid computational difficulties involving irreducibly high-dimensional integrals, we propose a Monte Carlo method based on the Metropolis algorithm for approximating the robust maximum likelihood estimators. We study the empirical properties of these estimators in simulations. We also illustrate the proposed robust method using clustered data on blood sugar content from a clinical trial of individuals who were investigated for diabetes.

Keywords

binary model censored regression model expectation maximization algorithm metropolis algorithm missing data robust estimation

Get full access to this article

View all access options for this article.

References

Wang

Bice

. A model with mixed binary responses and censored observations. Commun Stat – Theory Methods 1997; 26: 921–941.

MacDonald

Robinson

. Cautionary tails about arbitrary deletion of observations; or, throwing the variance out with the bathwater. J Labor Econ 1985; 3: 124–152.

Brown

. Protecting against nonrandomly missing data in longitudinal studies. Biometrics 1990; 46: 143–157.

Dantan

Proust-Lima

Letenneur

. Pattern mixture models and latent class models for the analysis of multivariate longitudinal data with informative dropouts. Int J Biostat 2008; 4: article 14–article 14.

Diggle

Kenward

. Informative dropout in longitudinal data analysis (with discussion). Appl Stat 1994; 43: 49–94.

Ibrahim

Lipsitz

Chen

. Missing covariates in generalized linear models when the missing data mechanism is non-ignorable. J R Stat Soc Ser B 1999; 61: 173–190.

Ibrahim

Chen

Lipsitz

. Missing responses in generalized linear mixed models when the missing data mechanism is nonignorable. Biometrika 2001; 88: 551–564.

Molenberghs

Verbeke

. A review on linear mixed models for longitudinal data, possibly subject to dropout. Stat Modell 2001; 1: 235–269.

Little

RJA

Rubin

. Statistical analysis with missing data, 2nd ed. New Jersey: John Wiley & Sons, 2002.

10.

Preisser

Galecki

Lohman

. Analysis of smoking trends with incomplete longitudinal binary responses. J Am Stat Assoc 2000; 95: 1021–1031.

11.

Sinha

. Robust methods for generalized linear models with nonignorable missing covariates. Can J Stat 2008; 36: 277–299.

12.

Verbeke

Molenberghs

. Longitudinal and incomplete clinical studies. Metron 2005; 63: 143–170.

13.

Liu

. A longitudinal study of children’s aggressive behaviours based on multivariate mixed models with incomplete data. Can J Stat 2009; 37: 435–452.

14.

Xie

. A local sensitivity analysis approach to longitudinal non-Gaussian data with non-ignorable dropout. Stat Med 2008; 27: 3155–3177.

15.

Cook

. Marginal methods for incomplete longitudinal data arising in clusters. J Am Stat Assoc 2002; 97: 1071–1080.

16.

McCulloch

. Maximum likelihood algorithms for generalized linear mixed models. J Am Stat Assoc 1997; 92: 162–170.

17.

Pregibon

. Resistant fits for some commonly used logistic models with medical applications. Biometrics 1982; 38: 485–498.

18.

Stefanski

Carroll

Ruppert

. Optimally bounded score functions for generalized linear models with applications to logistic regression. Biometrika 1986; 73: 413–424.

19.

Kunsch

Stefansky

Carroll

. Conditionally unbiased bounded-influence estimation in general regression models, with applications to generalized linear models. J Am Stat Assoc 1989; 84: 460–466.

20.

Morgenthaler

. Least-absolute-deviations fits for generalized linear models. Biometrika 1992; 79: 747–754.

21.

Carroll

Pederson

. On robustness in the logistic regression model. J R Stat Soc Ser B 1993; 55: 693–706.

22.

Preisser

Qaqish

. Robust regression to clustered data with application to binary responses. Biometrics 1999; 55: 574–579.

23.

Cantoni

Ronchetti

. Robust inference for generalized linear models. J Am Stat Assoc 2001; 96: 1022–1030.

24.

Sinha

. Robust analysis of generalized linear mixed models. J Am Stat Assoc 2004; 99: 451–460.

25.

Little

RJA

. Robust estimation of the mean and covariance matrix from data with missing values. Appl Stat 1988; 37: 23–38.

26.

Little

RJA

Smith

. Editing and imputation for quantitative survey data. J Am Stat Assoc 1987; 82: 58–68.

27.

McCulloch

. Maximum likelihood variance components estimation for binary data. J Am Stat Assoc 1994; 89: 330–335.