Abstract
A variable selection procedure is proposed using smooth-threshold generalized estimating equations based on quadratic inference functions (SGEE-QIF). The proposed procedure automatically eliminates inactive predictors by setting the corresponding parameters to be zero, and simultaneously estimates the nonzero regression coefficients by solving the SGEE-QIF. The proposed procedure avoids the convex optimization problem and is flexible and easy to implement. We establish the consistency and asymptotic normality of the resulting estimators. Extensive Monte Carlo simulation studies are conducted to examine the finite sample performance of the proposed variable selection procedure.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
References
Crowder, M.: On the use of a working correlation matrix in using generalised linear models for repeated measures. Biometrika 82, 407–410 (1995)
Dziak, J.J.: Penalized quadratic inference functions for variable selection in longitudinal research. Ph.D Thesis, The Pennsylvania State University, (2006), http://www.stat.psu.edu/~jdziak
Fan, J.Q., Li, R.: Variable selection via nonconcave penalized likelihood and its oracle properties. J. Amer. Statist. Assoc. 96, 1348–1360 (2001)
Frank, I.E., Friedman, J.H.: A statistical view of some chemometrics regression tools (with discussion). Technometrics 35, 109–148 (1993)
Hansen, L.: Large sample properties of generalized method of moments estimators. Econometrica 50, 1029–1054 (1982)
Lai, P., Wang, Q., Lian, H.: Bias-corrected GEE estimation and smooth-threshold GEE variable selection for single-index models with clustered data. J. Multivariate Anal. 105, 422–432 (2012)
Liang, K.L., Zeger, S.L.: Longitudinal data analysis using generalised estimating equations. Biometrika 73, 13–22 (1986)
Qu, A., Lindsay, B.G., Li, B.: Improving generalized estimating equations using quadratic inference functions. Biometrika 87, 823–836 (2000)
Qu, A., Song, P.X.K.: Assessing robustness of generalized estimating equations and quadratic inference functions. Biometrika 91, 447–459 (2004)
Tibshirani, R.: Regression shrinkage and selection via the Lasso. J. Roy. Statist. Soc., Ser. B 58, 267–288 (1996)
Ueki, M.: A note on automatic variable selection using smooth-threshold estimating equations. Biometrika 96(4), 1005–1011 (2009)
Xu, D., Zhang, Z.: Regularized REML for estimation in heteroscedastic regression models. In: Li, S., Wang, X., Okazaki, Y., Kawabe, J., Murofushi, T., Guan, L. (eds.) Nonlinear Mathematics for Uncertainty and its Applications. AISC, vol. 100, pp. 495–502. Springer, Heidelberg (2011)
Xu, D., Zhang, Z., Wu, L.: Variable selection in high-dimensional double generalized linear models. Stat. Papers (2012), doi:10.1007/s00362-012-0481-y
Xue, L., Qu, A., Zhou, J.J.: Consistent model selection for marginal generalized additive model for correlated data. J. Amer. Statist. Assoc. 105, 1518–1530 (2010)
Zhao, P.X., Xue, L.G.: Variable selection for semiparametric varying coefficient partially linear errors-in-variables models. J. Multivariate Anal. 101, 1872–1883 (2010)
Zhou, J., Qu, A.: Informative estimation and selection of correlation structure for longitudinal data. J. Amer. Statist. Assoc. 107, 701–710 (2012)
Zou, H.: The adaptive Lasso and its oracle properties. J. Amer. Statist. Assoc. 101, 1418–1429 (2006)
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2013 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Tian, R., Xue, L. (2013). Smooth-Threshold GEE Variable Selection Based on Quadratic Inference Functions with Longitudinal Data. In: Yang, Y., Ma, M., Liu, B. (eds) Information Computing and Applications. ICICA 2013. Communications in Computer and Information Science, vol 391. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-642-53932-9_10
Download citation
DOI: https://doi.org/10.1007/978-3-642-53932-9_10
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-53931-2
Online ISBN: 978-3-642-53932-9
eBook Packages: Computer ScienceComputer Science (R0)