Local influence for generalized linear mixed models. (English) Zbl 1042.62068
Summary: The authors describe a method for assessing model inadequacy in maximum likelihood estimation of a generalized linear mixed model. They treat the latent random effects in the model as missing data and develop the influence analysis on the basis of a \(Q\)-function which is associated with the conditional expectation of the complete-data log-likelihood function in the EM algorithm. They propose a procedure to detect influential observations in six model perturbation schemes. They also illustrate their methodology in a hypothetical situation and in two real cases.
MSC:
| 62J12 | Generalized linear models (logistic models) |
| 62J20 | Diagnostics, and linear inference and regression |
| 65C60 | Computational problems in statistics (MSC2010) |
Keywords:
generalized linear mixed model; local influence; missing data; \(Q\)-displacement function; perturbationsReferences:
| [1] | Beekman, Diagnostics for mixed model analysis of variance, Technometrics 29 pp 413– (1987) |
| [2] | Booth, Maximum generalized linear mixed model likelihoods with an automated Monte Carlo EM algorithm, Journal of the Royal Statistical Society Series B 61 pp 265– (1999) · Zbl 0917.62058 |
| [3] | Breslow, Extra-Poisson variation in log-linear models, Applied Statistics 33 pp 38– (1984) |
| [4] | Breslow, Approximate inference in generalized linear mixed models, Journal of the American Statistical Association 88 pp 9– (1993) · Zbl 0775.62195 |
| [5] | Cook, Assessment of local influence (with discussion), Journal of the Royal Statistical Society Series B 48 pp 133– (1986) · Zbl 0608.62041 |
| [6] | Crowder, Beta-binomial ANOVA for proportions, Applied Statistics 27 pp 34– (1978) |
| [7] | Davidian, Nonlinear Models for Repeated Measurement Data. (1995) |
| [8] | Gamerman, Sampling from the posterior distribution in generalized linear mixed models, Statistics and Computing 7 pp 57– (1997) |
| [9] | Gu, A stochastic approximation algorithm with Markov chain Monte-Carlo method for incomplete data estimation problems, Proceedings of the National Academy of Sciences of the United States of America 95 pp 7270– (1998) · Zbl 0898.62101 |
| [10] | Gu, Maximum likelihood estimation for spatial models by Markov chain Monte Carlo stochastic approximation, Journal of the Royal Statistical Society Series B 63 pp 339– (2001) · Zbl 0979.62060 |
| [11] | Hastings, Monte Carlo sampling methods using Markov chains and their applications, Biometrika 57 pp 97– (1970) · Zbl 0219.65008 |
| [12] | Lee, Sensitivity analysis of structural equation models, Psychometrika 61 pp 93– (1996) · Zbl 0866.62044 |
| [13] | Liu, Monte Carlo Strategies in Scientific Computing. (2001) · Zbl 0991.65001 |
| [14] | Manton, Empirical Bayes procedures for stabilizing maps of U.S. cancer mortality rates, Journal of the American Statistical Association 84 pp 637– (1989) |
| [15] | McCullagh, Generalized Linear Models (1989) · Zbl 0588.62104 · doi:10.1007/978-1-4899-3242-6 |
| [16] | McCulloch, Maximum likelihood variance components estimation for binary data, Journal of the American Statistical Association 89 pp 330– (1994) · Zbl 0800.62139 |
| [17] | McCulloch, Maximum likelihood algorithms for generalized linear mixed models, Journal of the American Statistical Association 92 pp 162– (1997) · Zbl 0889.62061 |
| [18] | McGilchrist, Estimation in generalized linear mixed models, Journal of the Royal Statistical Society Series B 56 pp 61– (1994) · Zbl 0800.62433 |
| [19] | Metropolis, Equations of state calculations by fast computing machine, Journal of Chemical Physics 21 pp 1087– (1953) |
| [20] | Mukerjee, On the positive definiteness of the information matrix under the binary and Poisson mixed models, Annals of the Institute of Statistical Mathematics 54 pp 355– (2002) · Zbl 1012.62076 |
| [21] | Poon, Conformal normal curvature and assessment of local influence, Journal of the Royal Statistical Society Series B 61 pp 51– (1999) · Zbl 0913.62062 |
| [22] | Poon, Influence analysis of structural equation models with polytomous variables, Psychometrika 64 pp 461– (1999) · Zbl 1365.62458 |
| [23] | Steele, A modified EM algorithm for estimation in generalized mixed models, Biometrics 52 pp 1295– (1996) · Zbl 0867.62060 |
| [24] | Stiratelli, Random effects models for serial observations with binary responses, Biometrics 40 pp 961– (1984) |
| [25] | Tanner, Tools for Statistical Inference: Methods for the Exploration of Posterior Distributions and Likelihood Functions (1996) · Zbl 0846.62001 |
| [26] | Thall, Some covariance models for longitudinal count data with overdispersion, Biometrics 46 pp 657– (1990) · Zbl 0712.62048 |
| [27] | Thomas, Assessing influence on regressing coefficients in generalized linear models, Biometrika 76 pp 741– (1989) |
| [28] | van Steen, A local influence approach to sensitivity analysis of incomplete longitudinal ordinal data, Statistical Modelling 1 pp 125– (2001) · Zbl 1022.62062 |
| [29] | Wei, Exponential Family Nonlinear Model. (1998) |
| [30] | Williams, Extra-binomial variation in logistic linear models, Applied Statistics 31 pp 144– (1982) · Zbl 0488.62055 |
| [31] | Zeger, Generalized linear models with random effects: a Gibbs sampling approach, Journal of the American Statistical Association 86 pp 79– (1991) |
| [32] | Zeger, Models for longitudinal data: a generalized estimating equation approach, Biometrics 44 pp 1049– (1988) · Zbl 0715.62136 |
| [33] | Zhu, Local influence for incomplete data models, Journal of the Royal Statistical Society Series B 63 pp 111– (2001) · Zbl 0976.62071 |
| [34] | Zhu, Analysis of generalized linear mixed models via a stochastic approximation algorithm with Markov chain Monte Carlo method, Statistics and Computing 12 pp 175– (2002) |
| [35] | Zhu, Case-deletion measures for models with incomplete data, Biometrika 88 pp 727– (2001) · Zbl 1006.62021 |
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.
