Abstract
Repeated measurement designs occur in many areas of statistical research. In 1986, Liang and Zeger offered an elegant analysis of these problems based on a set of generalized estimating equations (GEEs) for regression parameters, that specify only the relationship between the marginal mean of the response variable and covariates. Their solution is based on iterated reweighted least squares fitting. In this paper, we propose a rank-based fitting procedure that only involves substituting a norm based on a score function for the Euclidean norm used by Liang and Zeger. Our subsequent fitting, while also an iterated reweighted least squares solution to GEEs, is robust to outliers in response space and the weights can easily be adapted for robustness in factor space. As with the fitting of Liang and Zeger, our rank-based fitting utilizes a working covariance matrix. We prove that our estimators of the regression coefficients are asymptotically normal. The results of a simulation study show that the our proposed estimators are empirically efficient and valid. We illustrate our analysis on a real data set drawn from a hierarchical (three-way nested) design.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsReferences
Abebe, A., & McKean, J. W. (2007). Highly efficient nonlinear regression based on the Wilcoxon norm. In D. Umbach (Ed.), Festschrift in honor of Mir Masoom Ali on the occasion of his retirement (pp. 340–357).
Abebe, A., & McKean, J. W. (2013). Weighted Wilcoxon estimators in nonlinear regression. Australian & New Zealand Journal of Statistics, 55, 401–420.
Abebe, A., McKean, J. W., & Kloke, J. D. (2010). Iterated Reweighted Rank-Based Estimates for GEE Models. Technical report, Western Michigan University.
Bilgic, Y. K., & Susmann, H. (2013). rlme: an R package for rank-based estimation and prediction in random effects nested models. The R Journal, 5, 71–79.
Brunner, E., & Denker, M. (1994). Rank statistics under dependent observations and applications to factorial designs. Journal of Statistical Inference and Planning, 42, 353–378.
Chang, W., McKean, J. W., Naranjo, J. D., & Sheather, S. J. (1999). High breakdown rank-based regression. Journal of the American Statistical Association, 94, 205–219.
Hettmansperger, T. P., & McKean, J. W. (2011). Robust nonparametric statistical methods (2nd ed.). Boca Raton, FL: Chapman-Hall.
Hollander, M., & Wolfe, D. A. (1999). Nonparametric statistical methods (2nd ed.). New York: Wiley.
Jaeckel, L. A. (1972). Estimating regression coefficients by minimizing the dispersion of the residuals. Annals of Mathematical Statistics, 43, 1449–1458.
Jung, S.-H., & Ying, Z. (2003). Rank-based regression with repeated measurements data. Biometrika, 90, 732–740.
Kloke, J., & McKean, J. W. (2014). Nonparametric statistical methods using R. Boca raton, FL: Chapman-Hall.
Kloke, J., McKean, J. W., & Rashid, M. (2009). Rank-based estimation and associated inferences for linear models with cluster correlated errors. Journal of the American Statistical Association, 104, 384–390.
Koul, H. L., Sievers, G. L., & McKean, J. W. (1987). An estimator of the scale parameter for the rank analysis of linear models under general score functions. Scandinavian Journal of Statistics, 14, 131–141.
Liang, K.-Y., & Zeger, S. L. (1986). Longitudinal data analysis using generalized linear models. Biometrika, 73, 13–22.
McKean, J. W., Naranjo, J. D., & Sheather, S. J. (1996). Diagnostics to detect differences in robust fits of linear models. Computational Statistics, 11, 223–243.
Pinheiro, J., Bates, D., DebRoy, S., Sarkar, D., & R Core Team (2016). nlme: Linearand Nonlinear Mixed Effects Models. R package version 3.1-128. http://CRAN.R-project.org/package=nlme
Qaqish, B. F., & Preisser, J. S. (1999). Resistant fits for regression with correlated outcomes an estimating equations approach. Journal of Statistical Planning and Inference, 75, 415–431.
Sievers, G. L., & Abebe, A. (2004). Rank estimation of regression coefficients using iterated reweighted least squares, Journal of Statistical Computation and Simulation, 74, 821–831.
Wang, Y. -G., & Zhu, M. (2006). Rank-based regression for analysis of repeated measures. Biometrika, 93, 459–464.
Werner, C., & Brunner, E. (2007). Rank methods for the analysis of clustered data. Computational Statistics and Data Analysis, 51, 6041–5054.
West, B. T., Welch, K. B., & Gatecki, A. T. (2006). Linear mixed models: a practicalguide using statistical software. Boca Raton, FL: CRC Press
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Appendix
Appendix
Proof of Theorem 4.1.
Let \(\boldsymbol{\alpha }^{{\ast}}(\boldsymbol{\beta }) = \hat{\boldsymbol{\alpha }} (\boldsymbol{\beta },\hat{\phi }(\boldsymbol{\beta }))\). Let k ≥ 1 be arbitrary but fixed. For i = 1, …, K, let
We then write the GEERB estimating equations (4.17) in the compact form
The GEERB estimator \(\hat{\boldsymbol{\beta }}_{R}^{(k)}\) solves this equation.
Similar to Liang and Zeger (1986), we first expand \(K^{-1/2}\sum _{i=1}^{K}\mathbf{Z}_{i}(\boldsymbol{\beta },\boldsymbol{\alpha }^{{\ast}}(\boldsymbol{\beta }))\) in a Taylor series about the true parameter \(\boldsymbol{\beta }\) and evaluated at \(\hat{\boldsymbol{\beta }}_{R}^{(k)}\). By the chain rule, the gradient in this expansion is given by,
Because \(\hat{\boldsymbol{\beta }}_{R}^{(k)}\) solves Eq. (4.28), the Taylor expansion evaluated at \(\hat{\boldsymbol{\beta }}_{R}^{(k)}\) is
Solving for \(\sqrt{K}(\hat{\boldsymbol{\beta }}_{R}^{(k)} -\boldsymbol{\beta })\), we obtain
Secondly, we fix \(\boldsymbol{\beta }\) and expand \(K^{-1/2}\sum _{i=1}^{K}\mathbf{Z}_{i}(\boldsymbol{\beta },\boldsymbol{\alpha }^{{\ast}}(\boldsymbol{\beta }))\) about the true parameter \(\boldsymbol{\alpha }\) and evaluated at \(\boldsymbol{\alpha }^{{\ast}}\) to get
where the o p (1) term is due to regularity conditions which imply that the remainder term is \(\frac{1} {K}O_{p}(1)\). Note that the weights are evaluated at the true parameters in this expansion too.
Because \(\mathbf{Z}_{i}(\boldsymbol{\beta },\boldsymbol{\alpha })\) is evaluated at the true parameters, we use the notation given in (4.18). Letting h it T be the tth row the product \(\mathbf{D}_{i}^{T}\mathbf{V}_{i}^{-1/2}\), we then have
The second equality holds because the weights are evaluated at the true parameters.
By Assumptions [A.5] and [A.6], it follows from Theorem 3.1 of Brunner and Denker (1994) and the usual Cramer-Wold device that \(\frac{1} {\sqrt{K}}\sum _{i=1}^{K}Z_{i}(\boldsymbol{\beta },\boldsymbol{\alpha })\) is asymptotically normal with mean 0 and variance-covariance matrix
Note that the form of the variance-covariance matrix follows from (4.32) and the independence between subjects.
Returning to expression (4.31), from the assumptions we have C ∗ = O p (1). Because the scores can change value at only a finite number of points [e.g., Sect. 3.2.1 of Hettmansperger and McKean (2011)], we can write B ∗ as
Assuming Lindeberg-Feller conditions for the quantity in braces, as in (A.4), it follows, similar to (4.32) that \(\sqrt{K}\mathbf{B}^{{\ast}} = O_{p}(1)\), and, hence, B ∗ = o p (1).
To finish the proof, we need to consider the terms in expression (4.29). A simple derivation shows that
By assumption, C = O p (1). Since,
arguments similar to those above show that \(K^{-1}\sum _{i=1}^{K}\mathbf{B}_{i} = o_{p}(1)\). Hence,
This and the discussion around expressions (4.32) and (4.33) finish the proof of Theorem 4.1.
As a final note, the asymptotic representation of the estimator is
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Abebe, A., McKean, J.W., Kloke, J.D., Bilgic, Y.K. (2016). Iterated Reweighted Rank-Based Estimates for GEE Models. In: Liu, R., McKean, J. (eds) Robust Rank-Based and Nonparametric Methods. Springer Proceedings in Mathematics & Statistics, vol 168. Springer, Cham. https://doi.org/10.1007/978-3-319-39065-9_4
Download citation
DOI: https://doi.org/10.1007/978-3-319-39065-9_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-39063-5
Online ISBN: 978-3-319-39065-9
eBook Packages: Mathematics and StatisticsMathematics and Statistics (R0)