Comparison of hierarchical and marginal likelihood estimators for binary outcomes. (English) Zbl 1429.62331
Summary: Likelihood estimation in random-effect models is often complicated because the marginal likelihood involves an analytically intractable integral. Numerical integration such as Gauss-Hermite quadrature is an option, but is generally not recommended when the dimensionality of the integral is high. An alternative is the use of hierarchical likelihood, which avoids such burdensome numerical integration. These two approaches for fitting binary data are compared and the advantages of using the hierarchical likelihood are discussed. Random-effect models for binary outcomes and for bivariate binary-continuous outcomes are considered.
MSC:
| 62J12 | Generalized linear models (logistic models) |
Keywords:
binary data; Gauss-Hermite quadrature; hierarchical likelihood; joint model; marginal likelihood; random effectsSoftware:
GenStatReferences:
| [1] | Aitken, A. C., On Bernoulli’s numerical solution of algebraic equations, Proc. Roy. Soc. Edinburgh, 46, 289-305 (1926) · JFM 52.0098.05 |
| [2] | Bjrnstad, J. F., On the generalization of the likelihood function and likelihood principle, J. Amer. Statist. Assoc., 91, 791-806 (1996) · Zbl 0871.62006 |
| [3] | Booth, J.; Hobert, J., Maximizing generalized linear mixed model likelihoods with an automated Monte Carlo EM algorithm, J. Roy. Statist. Soc. B, 62, 265-285 (1999) · Zbl 0917.62058 |
| [4] | Breslow, N. E.; Clayton, D. G., Approximate inference in generalized linear mixed models, J. Amer. Statist. Assoc., 88, 9-25 (1993) · Zbl 0775.62195 |
| [5] | Breslow, N. E.; Lin, X., Bias correction in generalized linear mixed models with a single component of dispersion, Biometrika, 82, 81-91 (1995) · Zbl 0823.62059 |
| [6] | Cox, D. R.; Reid, N., Parameter orthogonality and approximate conditional inference, J. Roy. Statist. Soc. B, 32, 1-18 (1987) · Zbl 0616.62006 |
| [7] | Crouch, E. A.C.; Spiegelman, D., The evaluation of integrals of the form \(∫−∞+∞f(t) exp(−t^2) dt\) application to logistic-normal models, J. Amer. Statist. Assoc., 85, 464-469 (1990) · Zbl 0716.65137 |
| [8] | Gueorguieva, R. V., A multivariate generalized linear mixed model for joint modelling of clustered outcomes in the exponential family, Statist. Modelling, 1, 177-193 (2001) · Zbl 1106.62060 |
| [9] | Ha, I. D.; Lee, Y.; Song, J.-K., Hierarchical likelihood approach for frailty models, Biometrika, 88, 233-243 (2001) · Zbl 1033.62096 |
| [10] | Ha, I. D.; Lee, Y.; Song, J.-K., Hierarchical likelihood approach for mixed linear models with censored data, Lifetime Data Anal., 8, 163-176 (2002) · Zbl 1037.62100 |
| [11] | Hinde, J., Compound poisson regression models, (Gilchrist, R., GLIM 82 (1982), Springer: Springer New York), 109-121 · Zbl 0505.62050 |
| [12] | Johnson, R. A.; Wichern, D. W., Applied Multivariate Statistical Analysis (1992), Prentice-Hall: Prentice-Hall Englewood Cliffs, NJ · Zbl 0745.62050 |
| [13] | Lee, Y., Robust variance estimators for fixed-effect estimates with hierarchical likelihood, Statist. Comput., 12, 201-207 (2002) |
| [14] | Lee, Y.; Nelder, J. A., Hierarchical generalized linear models, J. Roy. Statist. Soc. B, 58, 619-678 (1996), (with discussion) · Zbl 0880.62076 |
| [15] | Lee, Y.; Nelder, J. A., Hierarchical generalized linear modelsa synthesis of generalized linear models, random-effect model and structured dispersion, Biometrika, 88, 987-1006 (2001) · Zbl 0995.62066 |
| [16] | Lee, Y.; Nelder, J. A., Modelling and analysing correlated non-normal data, Statist. Modelling, 1, 3-16 (2001) · Zbl 1004.62080 |
| [17] | Lee, Y.; Nelder, J. A., Analysis of the ulcer data using hierarchical generalized linear models, Statist. Med., 21, 191-202 (2002) |
| [18] | Lesaffre, E.; Spiessens, B., On the effect of the number of quadrature points in a logistic random-effects modelan example, Appl. Statist., 50, 325-335 (2001) · Zbl 1112.62307 |
| [19] | Liu, Q.; Pierce, D. A., Heterogeneity in Mantel-Haenszel-type models, Biometrika, 80, 543-546 (1993) · Zbl 0800.62702 |
| [20] | Longford, N. T., Logistic regression with random coefficients, Comput. Statist. Data. Anal., 49, 975-988 (1994) · Zbl 0937.62520 |
| [21] | McCulloch, C. E., Maximum likelihood algorithms for generalized linear mixed models, J. Amer. Statist. Assoc., 92, 162-170 (1997) · Zbl 0889.62061 |
| [22] | Pawitan, Y., In all likelihood: statistical modelling and inference using likelihood (2001), Clarendon Press: Clarendon Press Oxford · Zbl 1013.62001 |
| [23] | Payne, R. W.; Baird, D. B.; Cherry, M.; Gilmour, A. R.; Harding, S. A.; Lane, P. W.; Morgan, G. W.; Murray, D. A.; Soutar, D. M.; Thompson, R.; Todd, A. D.; Tunnicliffe Wilson, G.; Webster, R.; Welham, S. J., The Guide to GenStat, Part 2: Statistics (2002), VSN International: VSN International Oxford |
| [24] | Price, C. J.; Kimmel, C. A.; Tyle, R. W.; Marr, M. C., The developmental toxicity of Ethylene Glycol in rates and mice, Toxicol. Appl. Pharmacol., 81, 113-127 (1985) |
| [25] | Stiratelli, R.; Laird, N.; Ware, J., Random effect model for serial observations with binary response, Biometric, 40, 961-971 (1984) |
| [26] | Wolfinger, R.D., 1999. Fitting nonlinear mixed models with the new NLMIXED procedure. Paper 287, Proceedings of the 99 Joint Statistical Meetings.; Wolfinger, R.D., 1999. Fitting nonlinear mixed models with the new NLMIXED procedure. Paper 287, Proceedings of the 99 Joint Statistical Meetings. |
| [27] | Zeger, S. L.; Karim, M. R., Generalized linear models with random effectsa Gibbs sampling approach, J. Amer. Statist. Assoc., 86, 79-86 (1991) |
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.
