Fixed and Random Effects Selection in Mixed Effects Models

Overview

Journal Biometrics

Publisher Oxford University Press

Specialty Public Health

Date 2010 Jul 29

PMID 20662831

Citations 29

Authors

Joseph G Ibrahim

Hongtu Zhu

Ramon I Garcia

Ruixin Guo

Affiliations

Soon will be listed here.

Abstract

We consider selecting both fixed and random effects in a general class of mixed effects models using maximum penalized likelihood (MPL) estimation along with the smoothly clipped absolute deviation (SCAD) and adaptive least absolute shrinkage and selection operator (ALASSO) penalty functions. The MPL estimates are shown to possess consistency and sparsity properties and asymptotic normality. A model selection criterion, called the IC(Q) statistic, is proposed for selecting the penalty parameters (Ibrahim, Zhu, and Tang, 2008, Journal of the American Statistical Association 103, 1648-1658). The variable selection procedure based on IC(Q) is shown to consistently select important fixed and random effects. The methodology is very general and can be applied to numerous situations involving random effects, including generalized linear mixed models. Simulation studies and a real data set from a Yale infant growth study are used to illustrate the proposed methodology.

Citing Articles

Variable selection in modelling clustered data via within-cluster resampling.

Ye S, Yu T, Caroff D, Huang S, Zhang B, Wang R Can J Stat. 2025; 53(1).

PMID: 40040799 PMC: 11878247. DOI: 10.1002/cjs.11824.

Frequency and Predictors of Adolescent Worry for School Gun Violence In the United States: Findings from a Nationally Representative Study.

Kirkland D, Fu T, Herbenick D, Hensel D medRxiv. 2025; .

PMID: 40034766 PMC: 11875268. DOI: 10.1101/2025.02.18.25322472.

HighDimMixedModels.jl: Robust high-dimensional mixed-effects models across omics data.

Gorstein E, Aghdam R, Solis-Lemus C PLoS Comput Biol. 2025; 21(1):e1012143.

PMID: 39804942 PMC: 11761659. DOI: 10.1371/journal.pcbi.1012143.

glmmPen: High Dimensional Penalized Generalized Linear Mixed Models.

Heiling H, Rashid N, Li Q, Ibrahim J R J. 2024; 15(4):106-128.

PMID: 38818017 PMC: 11138212. DOI: 10.32614/rj-2023-086.

Efficient computation of high-dimensional penalized generalized linear mixed models by latent factor modeling of the random effects.

Heiling H, Rashid N, Li Q, Peng X, Yeh J, Ibrahim J Biometrics. 2024; 80(1).

PMID: 38497825 PMC: 10946237. DOI: 10.1093/biomtc/ujae016.

References

Daniels M, Kass R . Shrinkage estimators for covariance matrices. Biometrics. 2002; 57(4):1173-84. PMC: 2748251. DOI: 10.1111/j.0006-341x.2001.01173.x. View

Ni X, Zhang D, Zhang H . Variable selection for semiparametric mixed models in longitudinal studies. Biometrics. 2009; 66(1):79-88. PMC: 2875374. DOI: 10.1111/j.1541-0420.2009.01240.x. View

Hunter D, Li R . Variable Selection using MM Algorithms. Ann Stat. 2009; 33(4):1617-1642. PMC: 2674769. DOI: 10.1214/009053605000000200. View

Ibrahim J, Lipsitz S . Parameter estimation from incomplete data in binomial regression when the missing data mechanism is nonignorable. Biometrics. 1996; 52(3):1071-8. View

Wang H, Li R, Tsai C . Tuning parameter selectors for the smoothly clipped absolute deviation method. Biometrika. 2009; 94(3):553-568. PMC: 2663963. DOI: 10.1093/biomet/asm053. View

Ibrahim J, Zhu H, Tang N . Model Selection Criteria for Missing-Data Problems Using the EM Algorithm. J Am Stat Assoc. 2009; 103(484):1648-1658. PMC: 2728244. DOI: 10.1198/016214508000001057. View

Ibrahim J, Chen M, Lipsitz S . Monte Carlo EM for missing covariates in parametric regression models. Biometrics. 2001; 55(2):591-6. DOI: 10.1111/j.0006-341x.1999.00591.x. View

Cai J, Fan J, Li R, Zhou H . Variable selection for multivariate failure time data. Biometrika. 2009; 92(2):303-316. PMC: 2674767. DOI: 10.1093/biomet/92.2.303. View

Qu A, Li R . Quadratic inference functions for varying-coefficient models with longitudinal data. Biometrics. 2006; 62(2):379-91. PMC: 2680010. DOI: 10.1111/j.1541-0420.2005.00490.x. View

10.

Garcia R, Ibrahim J, Zhu H . Variable selection in the cox regression model with covariates missing at random. Biometrics. 2009; 66(1):97-104. PMC: 3303197. DOI: 10.1111/j.1541-0420.2009.01274.x. View

11.

Wasserman D, LEVENTHAL J . Maltreatment of children born to cocaine-dependent mothers. Am J Dis Child. 1993; 147(12):1324-8. DOI: 10.1001/archpedi.1993.02160360066021. View

12.

Zou H, Li R . One-step Sparse Estimates in Nonconcave Penalized Likelihood Models. Ann Stat. 2009; 36(4):1509-1533. PMC: 2759727. DOI: 10.1214/009053607000000802. View

13.

Stier D, LEVENTHAL J, Berg A, Johnson L, Mezger J . Are children born to young mothers at increased risk of maltreatment?. Pediatrics. 1993; 91(3):642-8. View

14.

Liang H, Wu H, Zou G . A Note on Conditional AIC for Linear Mixed-Effects Models. Biometrika. 2009; 95(3):773-778. PMC: 2572765. DOI: 10.1093/biomet/asn023. View

15.

Claeskens G, Consentino F . Variable selection with incomplete covariate data. Biometrics. 2008; 64(4):1062-9. DOI: 10.1111/j.1541-0420.2008.01003.x. View

16.

Garcia R, Ibrahim J, Zhu H . VARIABLE SELECTION FOR REGRESSION MODELS WITH MISSING DATA. Stat Sin. 2010; 20(1):149-165. PMC: 2844735. View

17.

Bondell H, Krishna A, Ghosh S . Joint variable selection for fixed and random effects in linear mixed-effects models. Biometrics. 2010; 66(4):1069-77. PMC: 2895687. DOI: 10.1111/j.1541-0420.2010.01391.x. View

18.

Breheny P, Huang J . Penalized methods for bi-level variable selection. Stat Interface. 2010; 2(3):369-380. PMC: 2904563. DOI: 10.4310/sii.2009.v2.n3.a10. View

19.

Thall P, Vail S . Some covariance models for longitudinal count data with overdispersion. Biometrics. 1990; 46(3):657-71. View

20.

Johnson B, Lin D, Zeng D . Penalized Estimating Functions and Variable Selection in Semiparametric Regression Models. J Am Stat Assoc. 2010; 103(482):672-680. PMC: 2850080. DOI: 10.1198/016214508000000184. View