Abstract
Linear mixed-effects models are widely used in analysis of longitudinal data. However, testing for zero-variance components of random effects has not been well resolved in statistical literature, although some likelihood-based procedures have been proposed and studied. In this article, we propose a generalized p-value based method in coupling with fiducial inference to tackle this problem. The proposed method is also applied to test linearity of the nonparametric functions in additive models. We provide theoretical justifications and develop an implementation algorithm for the proposed method. We evaluate its finite-sample performance and compare it with that of the restricted likelihood ratio test via simulation experiments. We illustrate the proposed approach using an application from a nutritional study.
Similar content being viewed by others
References
Arendacká B. Generalized confidence intervals on the variance component in mixed linear models with two variance components. Statistics, 2005, 39: 275–286
Brumback B A, Ruppert D, Wand M. Comment on "Variable selection and function estimation in additive nonparametric regression using a data-based prior" by Shively, Kohn and Wood. J Amer Statist Assoc, 1999, 94: 794–797
Crainiceanu C M, Ruppert D. Restricted likelihood ratio tests in nonparametric longitudinal models. Statist Sinica, 2004, 14: 713–729
Crainiceanu C M, Ruppert D. Likelihood ratio tests in linear mixed models with one variance component. J R Stat Soc Ser B Stat Methodol, 2004, 66: 165–185
Crainiceanu C M, Ruppert D, Claeskens G, et al. Likelihood ratio tests of polynomial regression against a general nonparametric alternative. Biometrika, 2005, 92: 91–103
Crainiceanu C M, Ruppert D, Vogelsang T. Some properties of likelihood ratio tests in linear mixed models. Technical report. New York: Cornell University, 2003
Das R, Sinha B K. Robust optimum invariant unbiased tests for variance components. In: Proceedings of the Second International Tampere Conference in Statistics. Tampere: University of Tampere, 1987, 317–342
David A P, Stone M. The functional model basis of fiducial inference. Ann Statist, 1982, 10: 1054–1067
Davidian M, Giltinan D M. Nonlinear Models for Repeated Measurement Data. New York: Chapman and Hall, 1995
El-Baddiouni M Y, Halawa A M. A combined invariant test for a null variance ratio. Biom J, 2003, 45: 249–259
Fairfield K M, Fletcher R H. Vitamins for chronic disease prevention in adults. J Amer Med Assoc, 2002, 287: 3116–3226
Fisher R A. Inverse probability. Math Proc Cambridge Philos Soc, 1930, 26: 528–535
Greven S, Crainiceanu C M, Kuechenhoff H, et al. Restricted likelihood ratio testing for zero variance components in linear mixed models. J Comput Graph Statist, 2008, 17: 870–891
Hannig J. On generalized fiducial inference. Statist Sinica, 2009, 19: 491–544
Hannig J, Iyer H. Fiducial intervals for variance components in an unbalanced two-component normal mixed linear model. J Amer Statist Assoc, 2008, 103: 854–865
Hannig J, Iyer H, Patterson P. Fiducial generalized confidence intervals. J Amer Statist Assoc, 2006, 101: 254–269
Härdle W, Müller M, Sperlich S, et al. Nonparametric and Semiparametric Models. Heidelberg: Springer-Verlag, 2004
Khuri A, Mathew T, Sinha B K. Statistical Tests for Mixed Linear Models. New York: John Wiley & Sons, 1998
Krishnamoorthy K, Mathew T. Inferences on the means of lognormal distributions using generalized p-values and generalized confidence intervals. J Statist Plann Inference, 2003, 115: 103–121
Laird N, Ware J. Random-effects models for longitudinal data. Biometrics, 1982, 38: 963–974
Mathew T, Webb D W. Generalized p-values and confidence intervals for variance components: Applications to army test and evaluation. Technometrics, 2005, 47: 312–322
Nierenberg D W, Stukel T A, Baron J A, et al. Determinants of plasma levels of beta-carotene and retinol. Amer J Epidemiol, 1989, 130: 511–521
Pinheiro J C, Bates D M. Mixed-Effects Models in S and S-PLUS. New York: Springer, 2000
Ruppert D, Wand M, Carroll R. Semiparametric Regression. New York: Cambridge University Press, 2003
Scheipl F, Greven S, Küchenhoff H. Size and power of tests for a zero random effect variance or polynomial regression in additive and linear mixed models. Comput Statist Data Anal, 2008, 52: 3283–3299
Self S, Liang K-Y. Asymptotic properties of maximum likelihood estimators and likelihood ratio tests under nonstandard conditions. J Amer Statist Assoc, 1987, 82: 605–610
Shephard N G. Maximum likelihood estimation of regression models with stochastic trend components. J Amer Statist Assoc, 1993, 88: 590–595
Stram D, Lee J-W. Variance components testing in the longitudinal mixed effects model. Biometrics, 1994, 50: 1171–1177
Tian L. Inferences on the within-subject coeffcient of variation. Stat Med, 2005, 25: 2008–2017
Tian L. Inferences on standardized mean difference: The generalized variable approach. Stat Med, 2006, 26: 945–953
Tsui K W, Weerahandi S. Generalized p-values in significance testing of hypotheses in the presence of nuisance parameters. J Amer Statist Assoc, 1989, 84: 602–607
Verbeke G, Molenberghs G. Linear Mixed Models for Longitudinal Data. New York: Springer, 2000
Vonesh E F, Chinchilli V M. Linear and Nonlinear Models for the Analysis of Repeated Measurements. New York: Marcel Dekker, 1996
Wald A. A note on regression analysis. Ann Math Statist, 1947, 18: 586–589
Weerahandi S. Testing variance components in mixed models with generalized p-values. J Amer Statist Assoc, 1991, 86: 151–153
Weerahandi S. Exact Statistical Methods for Data Analysis. New York: Springer-Verlag, 1995
Weerahandi S. Generalized Inference in Repeated Measures: Exact Methods in MANOVA and Mixed Models. New Jersey: John Wiley & Sons, 2004
Zhou L, Mathew T. Some tests for variance components using generalized p-values. Technometrics, 1994, 36: 394–402
Acknowledgements
This work was supported by Shandong Provincial Natural Science Foundation of China (Grant No. ZR2014AM019), National Natural Science Foundation of China (Grant Nos. 11171188 and 11529101), the Scientific Research Foundation for the Returned Overseas Chinese Scholars, State Education Ministry of China, and National Science Foundation of USA (Grant Nos. DMS-1418042 and DMS-1620898). The authors thank two reviewers for their valuable suggestions.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Li, X., Su, H. & Liang, H. Fiducial generalized p-values for testing zero-variance components in linear mixed-effects models. Sci. China Math. 61, 1303–1318 (2018). https://doi.org/10.1007/s11425-016-9068-8
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11425-016-9068-8
Keywords
- fiducial distribution
- generalized pivotal quantity
- generalized test variable
- penalized spline additive models
- restricted likelihood ratio test
- structural equation
- zero-variance components