失混合效应模型的分位回归及变量选择
Quantile Regression for Censored Mixed Effects Models and Variable Selection
纵向数据常常用正态混合效应模型进行分析.然而,违背正态性的假定往往会导致无效的推断.与传统的均值回归相比较,分位回归可以给出响应变量条件分布的完整刻画,对于非正态误差分布也可以给稳健的估计结果.本文主要考虑右删失响应下纵向混合效应模型的分位回归估计和变量选择问题.首先,逆删失概率加权方法被用来得到模型的参数估计.其次,结合逆删失概率加权和LASSO惩罚变量选择方法考虑了模型的变量选择问题.蒙特卡洛模拟显示所提方法要比直接删除删失数据的估计方法更具优势.最后,分析了一组艾滋病数据集来展示所提方法的实际应用效果.
Longitudinal data are usually analyzed using mixed effects models under the assumption of normal distributions. A departure from normality may result in invalid inference. Compared with the traditional mean regression, quantile regression can characterize a complete scan of the conditional distribution of the response variable and provide more robust inferences for nonnormal error distributions. In this paper, we focus on the quantile estimation and variable selection of censored mixed effects models. Firstly, the inverse censoring probability weighted (ICPW) method is utilized to obtain parameters estimation. Furthermore, the LASSO penalties are incorporated into the ICPW method to implement variable selection. Monte Carlo simulations demonstrate that the proposed method performs superior to the "naive" method which ignores censored data. Finally, an AIDS data set is analyzed to illustrate the proposed method.
分位回归 / 删失混合效应模型 / 逆删失概率加权方法 / 变量选择 / LASSO惩罚 {{custom_keyword}} /
quantile regression / censored mixed effects models / ICPW method / variable selection / LASSO penalty {{custom_keyword}} /
[1] Arellano M., Madrid C., Bonhomme S., et al., Random Effects Quantile Regression, Working Paper, 2013.
[2] Bang H., Tsiatis A. A., Median regression with censored cost data, Biometrics, 2002, 58:643-649.
[3] Bondell H. D., Krishna A., Ghosh S. K., Joint variable selection for fixed and random effects in linear mixed-effects models, Biometrics, 2010, 66:1069-1077.
[4] Chen Z., Dunson D. B., Random effects selection in linear mixed models, Biometrics, 2003, 59:762-769.
[5] Davino C., Furno M., Vistocco D., Quantile Regression:Theory and Applications, John Wiley & Sons, New York, 2014.
[6] Fan J., Li R., Variable selection via nonconcave penalized likelihood and its oracle properties, J. Amer. Statistical Association, 2001, 96:1348-1360.
[7] Fan Y., Li R., Variable selection in linear mixed effects models, The Annals of Statistics, 2012, 4:2043-2068.
[8] Fleming T. R., Harrington D. P., Counting Processes and Survival Analysis, Wiley, New York, 1991.
[9] Galvao A. F., Lamarche C., Lima L. R., Estimation of censored quantile regression for panel data with fixed effects, J. Amer. Statistical Association, 2013, 108(503):1075-1089.
[10] Geraci M., Bottai M., Quantile regression for longitudinal data using the asymmetric Laplace distribution, Biostatistics, 2007, 8:140-154.
[11] Gill R. D., Censoring and stochastic integrals, Statistica Neerlandica, 1980, 34(2):124-124.
[12] Ibrahim J. G., Zhu H., Garcia R. I., et al., Fixed and random effects selection in mixed effects models, Biometrics, 2011, 67:495-503.
[13] Jiang R., Qian W., Zhou Z., Variable selection and coefficient estimation via composite quantile regression with randomly censored data, Statistics & Probability Letters, 2012, 82:308-317.
[14] Kaplan E. L., Meier P., Nonparametric estimation from incomplete observations, J. Amer. Statistical Association, 1958, 53(282):457-481.
[15] Kim M. O., Yang Y., Semiparametric approach to a random effects quantile regression model, J. Amer. Statistical Association, 2011, 106(496):1405-1417.
[16] Knight K., Limiting distributions for L1 regression estimators under general conditions, The Annals of Statistics, 1998, 26:755-770.
[17] Kobayashi G., Kozumi H., Bayesian analysis of quantile regression for censored dynamic panel data, Computational Statistics, 2013, 27:359-380.
[18] Koenker R., Bassett J. G., Regression quantiles, Econometrica, 1978, 46(1):33-50.
[19] Koenker R., Quantile regression for longitudinal data, J. Multivariate Analysis, 2004, 91(1):74-89.
[20] Koenker R., Quantile Regression, Cambridge University Press, London, 2005.
[21] Lachos V. H., Chen M. H., Abanto-Vallec C. A., et al., Quantile Regression for Censored Mixed-effects Models with Applications to HIV Studies, Statistics & Its Interface, 2015, 8(2):203-215.
[22] Laird N. M., Ware J. H., Random-effects models for longitudinal data, Biometrics, 1982, 38(4):963-74.
[23] Lederman M. M., Connick E., Landay A., et al., Immunologic responses associated with 12 weeks of combination antiretroviral therapy consisting of zidovudine, lamivudine, and ritonavir:results of AIDS Clinical Trials Group Protocol 315, J. Infectious Diseases, 1998, 178(1):70-79.
[24] Li G. R., Lai P., Lian H., Variable selection and estimation for partially linear single-index models with longitudinal data, Statistics and Computing, 2015, 25(3):579-593.
[25] Li G. R., Lian H., Feng S. Y. et al., Automatic variable selection for longitudinal generalized linear models, Computational Statistics and Data Analysis, 2013, 61:174-186.
[26] Li G. R., Lian H., Lai P., et al., Variable selection for fixed effects varying coefficient models, Acta Math. Sin., Engl. Ser., 2015, 31(1):91-110.
[27] Lin X., Variance component testing in generalised linear models with random effects, Biometrika, 1997, 84:309-326.
[28] Liu Y., Bottai M., Mixed-effects models for conditional quantiles with longitudinal data, International J. Biostatistics, 2009, 5(1):28-28.
[29] Maitre P. O., Buhrer M., Thomson D., et al., A three-step approach combining Bayesian regression and NONMEM population analysis:application to midazolam, J. Pharmacokinetics and Biopharmaceutics, 1991, 19:377-385.
[30] Mandema J. W., Verotta D., Sheiner L. B., Building population pharmacokinetic pharmacodynamic models, J. Pharmacokinetics and Biopharmaceutics, 1992, 20:511-529.
[31] Pakker N. G., Notermans D. W, De Boer R. J., et al., Biphasic kinetics of peripheral blood T cells after triple combination therapy in HIV-1 infection:a composite of redistribution and proliferation, Nature Medicine, 1998, 4:208-214.
[32] Peng L., Huang Y., Survival analysis with quantile regression models, J. Amer. Statistical Association, 2008, 103(482):637-649.
[33] Portnoy S., Censored regression quantiles, J. Amer. Statistical Association, 2003, 98:1001-1012.
[34] Pu W., Niu X., Selecting mixed-effects models based on a generalized information criterion, J. Multivariate Analysis, 2006, 97:733-758.
[35] Shows H., Lu W., Zhang H., Sparse estimation and inference for censored median regression, J. Planning and Inference, 2010, 140:1903-1917.
[36] Tang L. J., Zhou Z. G., Wu C. C., Weighted composite quantile estimation and variable selection method for censored regression model, Statistics and Probability Letters, 2012, 82:653-663.
[37] Tibshirani R., Regression shrinkage and selection via the Lasso, J. Royal Statistical Society, Series B, 1996, 58:267-288.
[38] Waldmann E., Kneib T., Yue Y. R., et al., Bayesian semiparametric additive quantile regression, Statistics & Probability, 2013, 13(3):223-252.
[39] Wang H. J., Fygenson M., Inference for censored quantile regression models in longitudinal studies, The Annals of Statistics, 2009, 37(2):756-781.
[40] Wang H., Wang L., Locally weighted censored quantille regression, J. Amer. Statistical Association, 2009, 104:1117-1128.
[41] Wang H., Zhou J., Li Y., Variable selection for censored quantile regression, Statistica Sinica, 2013, 23(1):145-167.
[42] Wu H., Kuritzkes D. R., McClernon D. R., et al., Characterization of viral dynamics in Human immunedeficiency virus type 1-infected patients treated with combination antiretroviral therapy:relationships to host factors, cellular restoration and virological endpoints, J. Infectious Diseases, 1999, 179:799-807.
[43] Wu H., Wu L., Identification of significant host factors for HIV dynamics modeled by nonlinear mixed-effect models, Statistics in Medicine, 2002, 21:753-771.
[44] Wu H., Zhang J. T., Nonparametric Regression Methods for Longitudinal Data Analysis:Mixed-effects Modeling Approaches, John Wiley & Sons, New York, 2006.
[45] Wu L., Mixed Effects Models for Complex Data, Chapman & Hall/CRC Press, Boca Raton, 2010.
[46] Wu L., Wu H., Nonlinear mixed-effect models with missing time-dependent covariates, with application to HIV Viral Dynamics, J. Royal Statistical Society, Series C (Applied Statistics), 2002, 51:297-318.
[47] Yang S., Censored median regression using weighted empirical survival and hazard functions, J. Amer. Statistical Association, 1999, 94(445):137-145.
[48] Ying Z., Jung S. H., Wei L. J., Survival analysis with median regression models, J. Amer. Statistical Association, 1995, 90(429):178-184.
[49] Zhou L., A simple censored median regression estimators, Statistica Sinica, 2006, 16:1043-1058.
[50] Zou H., The adaptive LASSO and its oracle properties, J. Amer. Statistical Association, 2006, 101:1418-1429.
国家自然科学基金资助项目(11501167,11271368);河南省高校重点科研基金资助项目(15A110025)
/
〈 | 〉 |