Bayesian quantile regression and variable selection for partial linear single-index model: Using free knot spline

ABSTRACT Partial linear single-index model (PLSIM) has both the flexibility of nonparametric treatment and interpretability of linear term, yet existing literatures about it mainly focused on mean regression, and quantile regression analysis is scarce. Based on free knot spline approximation, we apply asymmetric Laplace distribution to implement Bayesian quantile regression, and perform variable selection in linear term and index vector via binary indicators. Our approach is exempt from regularity conditions in frequentist method, and could execute variable selection and quantile regression under mutual posterior correction, which is also the first work to implement them jointly for PLSIM in fully Bayesian framework. The numerical simulation manifests the superiority of our approach to previous methods, which embodied in better efficiency of variable selection, index vector estimates and link function approximation with different error distributions. For illustration of its application, we build a power consumption model of A2/O process in wastewater treatment and emphatically analyze the impact of water quality factors.

[1]  Hai-Bin Wang,et al.  Bayesian estimation and variable selection for single index models , 2009, Comput. Stat. Data Anal..

[2]  Adrian F. M. Smith,et al.  Automatic Bayesian curve fitting , 1998 .

[3]  Peng Shi,et al.  A Bayesian quantile regression model for insurance company costs data , 2016 .

[4]  Jianqing Fan,et al.  Generalized Partially Linear Single-Index Models , 1997 .

[5]  H. Zou.  The Adaptive Lasso and Its Oracle Properties , 2006 .

[6]  Getachew A Dagne,et al.  Bayesian semiparametric zero-inflated Poisson model for longitudinal count data. , 2010, Mathematical biosciences.

[7]  Rahim Alhamzawi,et al.  Bayesian Elastic Net Tobit Quantile Regression , 2016, Commun. Stat. Simul. Comput..

[8]  Taeryon Choi,et al.  On asymptotic properties of Bayesian partially linear models , 2013 .

[9]  Nian-Sheng Tang,et al.  A semiparametric Bayesian approach to generalized partial linear mixed models for longitudinal data , 2012, Comput. Stat. Data Anal..

[10]  Jie Guo,et al.  Variable selection in high-dimensional partially linear additive models for composite quantile regression , 2013, Comput. Stat. Data Anal..

[11]  Hu Yang,et al.  A robust and efficient estimation and variable selection method for partially linear single-index models , 2014, J. Multivar. Anal..

[12]  Taeryon Choi,et al.  Asymptotic properties of posterior distributions in nonparametric regression with non-Gaussian errors , 2009 .

[13]  A. Gelfand,et al.  Spatial Quantile Multiple Regression Using the Asymmetric Laplace Process , 2012 .

[14]  Rahim Alhamzawi,et al.  Conjugate priors and variable selection for Bayesian quantile regression , 2013, Comput. Stat. Data Anal..

[15]  Hai-Bin Wang,et al.  Bayesian analysis of generalized partially linear single-index models , 2013, Comput. Stat. Data Anal..

[16]  Mary J. Lindstrom,et al.  Bayesian estimation of free-knot splines using reversible jumps , 2002, Comput. Stat. Data Anal..

[17]  Jane-ling Wang,et al.  Estimation for a partial-linear single-index model , 2009, 0905.2042.

[18]  Jing Yang,et al.  The adaptive L1-penalized LAD regression for partially linear single-index models , 2014 .

[19]  Yves F. Atchadé,et al.  A computational framework for empirical Bayes inference , 2011, Stat. Comput..

[20]  Rahim Alhamzawi,et al.  Prior elicitation and variable selection for bayesian quantile regression , 2013 .

[21]  R. O’Hara,et al.  A review of Bayesian variable selection methods: what, how and which , 2009 .

[22]  Rahim Alhamzawi,et al.  Bayesian lasso binary quantile regression , 2013, Comput. Stat..

[23]  T. Lancaster,et al.  Bayesian Quantile Regression , 2005 .

[24]  Kazushi Ikeda,et al.  An efficient sampling algorithm with adaptations for Bayesian variable selection , 2015, Neural Networks.

[25]  R. Koenker,et al.  Goodness of Fit and Related Inference Processes for Quantile Regression , 1999 .

[26]  Ruibin Xi,et al.  Bayesian regularized quantile regression , 2010 .

[27]  Krzysztof Podgórski,et al.  A Multivariate and Asymmetric Generalization of Laplace Distribution , 2000, Comput. Stat..

[28]  Runze Li,et al.  ESTIMATION AND TESTING FOR PARTIALLY LINEAR SINGLE-INDEX MODELS. , 2010, Annals of statistics.

[29]  P. Green.  Reversible jump Markov chain Monte Carlo computation and Bayesian model determination , 1995 .

[30]  B. Mallick.  VARIABLE SELECTION FOR REGRESSION MODELS , 2016 .

[31]  Rahim Alhamzawi,et al.  Bayesian model selection in ordinal quantile regression , 2016, Comput. Stat. Data Anal..

[32]  D. Ruppert,et al.  Penalized Spline Estimation for Partially Linear Single-Index Models , 2002 .

[33]  R. Koenker,et al.  Regression Quantiles , 2007 .

[34]  M. Schervish,et al.  On posterior consistency in nonparametric regression problems , 2007 .

[35]  Jianqing Fan,et al.  Variable Selection via Nonconcave Penalized Likelihood and its Oracle Properties , 2001 .

[36]  Weihua Zhao,et al.  Quantile regression and variable selection of partial linear single-index model , 2014, Annals of the Institute of Statistical Mathematics.

[37]  Keming Yu,et al.  Variable selection in quantile regression via Gibbs sampling , 2012 .

[38]  Yi-qiang,et al.  Variable Selection of Partially Linear Single-index Models , 2014 .

[39]  Graciela Boente,et al.  Robust estimates in generalized partially linear single-index models , 2011, TEST.