Electronic Journal of Statistics

A note on BIC in mixed-effects models

Maud Delattre, Marc Lavielle, and Marie-Anne Poursat

Full-text: Open access


The Bayesian Information Criterion (BIC) is widely used for variable selection in mixed effects models. However, its expression is unclear in typical situations of mixed effects models, where simple definition of the sample size is not meaningful. We derive an appropriate BIC expression that is consistent with the random effect structure of the mixed effects model. We illustrate the behavior of the proposed criterion through a simulation experiment and a case study and we recommend its use as an alternative to various existing BIC versions that are implemented in available software.

Article information

Electron. J. Statist., Volume 8, Number 1 (2014), 456-475.

First available in Project Euclid: 2 May 2014

Permanent link to this document

Digital Object Identifier

Mathematical Reviews number (MathSciNet)

Zentralblatt MATH identifier

Primary: 62J02: General nonlinear regression 62J12: Generalized linear models

Bayesian Information Criterion BIC mixed effects model variable selection


Delattre, Maud; Lavielle, Marc; Poursat, Marie-Anne. A note on BIC in mixed-effects models. Electron. J. Statist. 8 (2014), no. 1, 456--475. doi:10.1214/14-EJS890. https://projecteuclid.org/euclid.ejs/1399035845

Export citation


  • [1] Bondell, H., Krishna, A., and Ghosh, S. (2010). Joint variable selection for fixed and random effects in linear mixed-effects models. Biometrics, 66 1069–1077.
  • [2] Comets, E., Lavenu, A., and Lavielle, M. (2011). saemix: stochastic approximation expectation maximization SAEM algorithm. R package version 0.96.
  • [3] Davidian, M. and Giltinan, D. (2003). Nonlinear models for repeated measurement data: an overview and update. Journal of Agricultural, Biological, and Environmental Statistics, 8 387–419.
  • [4] Delattre, M. and Lavielle, M. (2012). Maximum likelihood estimation in discrete mixed hidden Markov models using the SAEM algorithm. Computational Statistics & Data Analysis, 56(6) 2073–2085.
  • [5] Delattre, M. and Lavielle, M. (2013). Coupling the SAEM algorithm and the extended Kalman filter for maximum likelihood estimation in mixed-effects diffusion models. Statistics and its interface, 6(4) 519–532.
  • [6] Delattre, M. (2012). Inférence statistique dans les modèles mixtes à dynamique Markovienne. PhD thesis, Université Paris-Sud.
  • [7] Delyon, B., Lavielle, M., and Moulines, E. (1999). Convergence of stochastic approximation version of the EM algorithm. Annals of Statistics, 27 94–128.
  • [8] Donohue, M., Overholser, R., Xu, R., and Vaida, F. (2011). Conditional Akaike information under generalized linear and proportional hazards mixed models. Biometrika, 98 685–700.
  • [9] Jiang, J. and Sunil Rao, J. (2003). Consistent procedures for mixed linear model selection. Sankhyã: The Indian Journal of Statistics, 65 23–42.
  • [10] Jiang, J., Sunil Rao, J., Gu, Z., and Nguyen, T. (2008). Fence methods for mixed model selection. The Annals of Statistics, 36 1669–1692.
  • [11] Kuhn, E. and Lavielle, M. (2005). Maximum likelihood estimation in nonlinear mixed-effects models. Computational Statistics and Data Analysis, 49 1020–1038.
  • [12] Lai, R., Huang, H., and Lee, T. (2012). Fixed and random effects selection in nonparametric additive mixed models. Electronic Journal of Statistics, 6 810–842.
  • [13] Laird, N. and Ware, J. (1982). Random-effects models for longitudinal data. Biometrics, 38 963–974.
  • [14] Lebarbier, E. and Mary-Huard, T. (2006). Une introduction au critère BIC: Fondements théoriques et interprétation. Journal de la Société Française de Statistique, 147(1) 39–57.
  • [15] Lian, H. (2012). A note on conditional Akaike information for Poisson regression with random effects. Electronic Journal of Statistics, 6 1–9.
  • [16] Liang, H., Wu, H., and Zou, G. (2008). A note on conditional AIC for linear mixed-effects models. Biometrika, 95 773–778.
  • [17] McCulloch, C. and Searle, S. (2008). Generalized, Linear, and Mixed Models. 2nd ed., Wiley, New-York.
  • [18] http://www.lixoft.com/monolix (2013). Monolix 4.2.2 User’s Guide.
  • [19] Nie, L. (2006). Strong consistency of the maximum likelihood estimator in generalized linear and nonlinear mixed-effects models. Metrika, 63 123–143.
  • [20] Nie, L. (2007). Convergence rate of the MLE in generalized linear and nonlinear mixed-effects models: Theory and applications. Journal of Statistical Planning and Inference, 137 1787–1804.
  • [21] O’Reilly, R. A. and Aggeler, P. M. (1968). Studies on coumarin anticoagulant drugs initiation of warfarin therapy without a loading dose. Circulation, 38(1) 169–177.
  • [22] Pinheiro, J. and Bates, D. (2009). Mixed-Effects Models in S and S-PLUS. 1st ed., 2nd printing, Springer-Verlag, New York Inc.
  • [23] Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., and R Core Team (2012). nlme: linear and lonlinear mixed effects models. R package version 3.1-105.
  • [24] Samson, A., Lavielle, M., and Mentré, F. (2007). The SAEM algorithm for group comparison tests in longitudinal analysis based on non-linear mixed-effects models. Statistics in medecine, 26 4860–4875.
  • [25] SAS (2008). SAS/STAT 9.2 User’s Guide, chapter 61, 4337–4435.
  • [26] Schelldorfer, J., Meier, L., and Bühlmann, P. (2013). GLMMLasso: An algorithm for high-dimensional generalized linear mixed models using $l_{1}$-penalization. Journal of Computational and Graphical Statistics, DOI:10.1080/10618600.2013.773239.
  • [27] Schwarz, G. (1978). Estimating the dimension of a model. Annals of Statistics, 6 461–464.
  • [28] SPSS (2002). Linear mixed-effects modeling in SPSS. An introduction to the MIXED procedure. Technical Report.
  • [29] Vaida, F. and Blanchard, S. (2005). Conditional Akaike information for mixed-effects models. Biometrika, 92 351–370.
  • [30] Vonesh, E. (1996). A note on the use of Laplace’s approximation for non-linear mixed-effects models. Biometrika, 83(2) 447–452.