Abstract
This paper is concerned with model selection in spline-based generalized linear mixed model. Exploiting the fact that smoothing parameters can be expressed as the reciprocal ratio of the variances of random effect under the setting of estimation by regularization, we propose a computationally efficient model selection procedure. Applications to some real data sets reveal that the proposed method selects reasonable models and is very fast to implement.
Similar content being viewed by others
References
Fitzmaurice GM, Laird NM, Ware JW (2004) Applied longitudinal analysis. Wiley-Interscience, Hoboken
Hastie T, Tibshirani R (1990) Generalized additive models. Chapman & Hall, New York
Hastie T, Tibshirani R, Friedman J (2001) The elements of statistical learning. Springer, New York
Kanba M, Yoshida T, Naito K (2008) On an efficient model selection in a spline mixed model, In: The proceeding of the 2008 IASC–ARS special conference, pp 799–808
Konishi S, Kitagawa G (1996) Generalised information criteria in model selection. Biometrika 83: 875–880
Konishi S, Kitagawa G (2008) Information criteria and statistical modeling. Springer, New York
Ruppert D, Wand MP, Carroll RJ (2003) Semiparametric regression. Cambridge University Press, Cambridge
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Yoshida, T., Kanba, M. & Naito, K. A computationally efficient model selection in the generalized linear mixed model. Comput Stat 25, 463–484 (2010). https://doi.org/10.1007/s00180-010-0187-3
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s00180-010-0187-3