Abstract
We study a linear mixed effects model for longitudinal data, where the response variable and covariates with fixed effects are subject to measurement error. We propose a method of moment estimation that does not require any assumption on the functional forms of the distributions of random effects and other random errors in the model. For a classical measurement error model we apply the instrumental variable approach to ensure identifiability of the parameters. Our methodology, without instrumental variables, can be applied to Berkson measurement errors. Using simulation studies, we investigate the finite sample performances of the estimators and show the impact of measurement error on the covariates and the response on the estimation procedure. The results show that our method performs quite satisfactory, especially for the fixed effects with measurement error (even under misspecification of measurement error model). This method is applied to a real data example of a large birth and child cohort study.
Similar content being viewed by others
References
Abarin T, Wu Y, Warrington N, Pennell C, Briollais L (2011) The impact of breastfeeding on FTO-related BMI growth trajectories: an application to the RAINE birth and child cohort study. Int J Epidemiol (to appear)
Abarin T, Wang L (2010) Instrumental variable approach to covariate measurement error in generalized linear models. Ann Inst Stat Math. doi:10.1007/s10463-010-0319-0
Abarin T, Wang L (2009) Second-order least squares estimation of censored regression models. J Stat Plan Inference 139:125–135
Abarin T, Wang L (2006) Comparison of GMM with second-order least squares estimation in nonlinear models. Far East J Theor Stat 20(2):179–196
Berkson J (1950) Are there two regressions? J Am Stat Assoc 45:164–180
Biggs R, et al. (2009) Spurious certainty: how ignoring measurement error and environmental heterogeneity may contribute to environmental controversies. Bioscience 59(1):65–76
Bland RM, et al. (2003) Maternal recall of exclusive breast feeding duration. Arch Dis Child 88(9):778–783
Buonaccorsi J (2010) Measurement error: models, methods, and applications. Chapman & Hall, London
Buonaccorsi J (1996) Measurement error in the response in the general linear model. J Am Stat Assoc 91:633–642
Buonaccorsi J, Tosteson T (1993) Correcting for nonlinear measurement errors in the dependent variable in the general linear model. Commun Stat, Theory Methods 22:2687–2702
Buonaccorsi J, Lin Ch (2002) Berkson measurement error in designed repeated measures studies with random coefficients. J Stat Plan Inference 104:53–72
Buonaccorsi J, et al. (2000) Estimation in longitudinal random effects models with measurement error. Stat Sin 10:885–903
Burton PR, et al. (2009) Size matters: just how big is BIG? Quantifying realistic sample size requirements for human genome epidemiology. Int J Epidemiol 38:263–273
Carroll RJ, et al. (2006) Measurement error in nonlinear models: a modern perspective, 2nd edn. Chapman & Hall, London
Carroll RJ, Wand MP (1991) Semiparametric estimation in logistic measurement error models. J R Stat Soc B 53:573–585
Faith MS, et al. (2004) Parental feeding attitudes and styles and child body mass index: prospective analysis of a gene-environment interaction. Pediatrics 114(4):e429–e436
Fuller AW (1987) Measurement error models. Wiley, New York
Joseph ML, Carriquiry A (2010) A measurement error approach to assess the association between dietary diversity, nutrient intake, and mean probability of adequacy. J Nutr 140(11):2094S–2101S
Kamarainen AM, et al. (2008) Zooplankton and the total phosphorus—chlorophyll a relationship: hierarchical Bayesian analysis of measurement error. Can J Fish Aquat Sci 65(12):2644–2655
Kipnis V, et al. (2003) Structure of dietary measurement error: results of the OPEN biomarker study. Am J Epidemiol 158(1):14–21
Laird NM, Ware JH (1982) Random-effects models for longitudinal data. Biometrics 38:963–974
Li H (2005) A simulation study of the second-order least squares estimators for nonlinear mixed effects models. Master’s thesis, University of Manitoba
Li H, Wang L (2011) Second-order least squares estimation in linear mixed models. University of Manitoba. Comm Statist Theory Methods doi:10.1080/s03610926.2011.601837
Marchand L, Wilkens L (2008) Design considerations for genomic association studies: importance of gene-environment interactions. Cancer Epidemiol Biomark Prev 17:263–267
Pan W, et al. (2009) Semiparametric transition measurement error models for longitudinal data. Biometrics 65:728–736
Parsons TJ, et al. (2001) Fetal and early life growth and body mass index from birth to early adulthood in 1958 British cohort: longitudinal study. Br Med J 323:1331–1335
Prentice RL, et al. (2002) Research strategies and the use of nutriet biomarkers in studies of diet and chronic disease. Public Health Nutr 5:977–984
Rampersaud E, et al. (2008) Physical activity and the association of common FTO gene variants with body mass index and obesity. Arch Intern Med 168(16):1791–1797
Rios E, et al. (1992) Accuracy of mothers’ responses to questions about breast-feeding practices. Food Nutr Bull 14(2):115–118
Schennach MS (2007) Instrumental variable estimation of nonlinear errors-in-variables models. Econometrica 75:201–239
Thomas D (2010) Methods for investigating gene-environment interactions in candidate pathway and genome-wide association studies. Annu Rev Public Health 31:21–36
Tosteson TD, et al. (1998) Covariate measurement error and the estimation of random effect parameters in a mixed model for longitudinal data. Stat Med 8:1139–1147
Tsiatis T, Davidian M (2001) A semiparametric estimator for the proportional hazards model with longitudinal covariates measured with error. Biometrika 88(2):447–458
Vimaleswaran KS, et al. (2009) Physical activity attenuates the body mass index–increasing influence of genetic variation in the FTO gene. Am J Clin Nutr 90(2):425–428
Wang L (2004) Estimation of nonlinear models with Berkson measurement errors. Ann Stat 32:2559–2579
Wang L, Hsiao C (2010) Method of moments estimation and identifiability of nonlinear semiparametric errors-in-variables models. J Econom 165:30–44. doi:10.1016/j.jeconom.2011.05.004
Wang N, et al. (1998) Bias analysis and SIMEX approach in generalized mixed measurement error models. J Am Stat Assoc 93:249–261
Wang N, et al. (1999) A bias correction regression calibration approach in generalized linear mixed measurement error model. Commun Stat, Theory Methods 28:217–232
Wong MY, et al. (2003) The detection of gene-environment interaction for continuous traits: should we deal with measurement error by bigger studies or better measurement? Int J Epidemiol 32:51–57
Acknowledgements
We thank the referees for their comments and helpful suggestions for manuscript changes. These have improved both the content and clarity of the manuscript. We would also like to thank Wei Ang, Nicole Warrington, Julie Marsh and Louise Mckenzie, for their help in preparing the data set and choosing the variables of the model in the application, and the RAINE study (http://www.rainestudy.org.au/) for providing the data for analysis. Financial support from “The Alva Foundation” and “Samuel Lunenfeld Research Institute—New Opportunities Funds” and MITACS are gratefully acknowledged.
Author information
Authors and Affiliations
Corresponding author
Additional information
This study has been partially supported by NSERC.
Electronic Supplementary Material
Below is the link to the electronic supplementary material.
Rights and permissions
About this article
Cite this article
Abarin, T., Li, H., Wang, L. et al. On Method of Moments Estimation in Linear Mixed Effects Models with Measurement Error on Covariates and Response with Application to a Longitudinal Study of Gene-Environment Interaction. Stat Biosci 6, 1–18 (2014). https://doi.org/10.1007/s12561-012-9074-5
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12561-012-9074-5