Single-index composite quantile regression for ultra-high-dimensional data

Composite quantile regression (CQR) is a robust and efficient estimation method. This paper studies CQR method for single-index models with ultra-high-dimensional data. We propose a penalized CQR estimator for single-index models and combine the debiasing technique with the CQR method to construct an estimator that is asymptotically normal, which enables the construction of valid confidence intervals and hypothesis testing. Both simulations and data analysis are conducted to illustrate the finite sample performance of the proposed methods.

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

[2]  H. Zou,et al.  Composite quantile regression and the oracle Model Selection Theory , 2008, 0806.2905.

[3]  Han Liu,et al.  A General Framework for Robust Testing and Confidence Regions in High-Dimensional Quantile Regression , 2014, 1412.8724.

[4]  S. Geer,et al.  On asymptotically optimal confidence regions and tests for high-dimensional models , 2013, 1303.0518.

[5]  Kengo Kato,et al.  Robust inference in high-dimensional approximately sparse quantile regression models , 2013 .

[6]  Heng Lian,et al.  A debiased distributed estimation for sparse partially linear models in diverging dimensions , 2017 .

[7]  R. Tibshirani.  Regression Shrinkage and Selection via the Lasso , 1996 .

[8]  T. Cai,et al.  A Constrained ℓ1 Minimization Approach to Sparse Precision Matrix Estimation , 2011, 1102.2233.

[9]  Tianxi Cai,et al.  L1-Regularized Least Squares for Support Recovery of High Dimensional Single Index Models with Gaussian Designs , 2015, J. Mach. Learn. Res..

[10]  Runze Li,et al.  NEW EFFICIENT ESTIMATION AND VARIABLE SELECTION METHODS FOR SEMIPARAMETRIC VARYING-COEFFICIENT PARTIALLY LINEAR MODELS. , 2011, Annals of statistics.

[11]  Gerda Claeskens,et al.  Confidence intervals for high-dimensional partially linear single-index models , 2016, J. Multivar. Anal..

[12]  Heng Lian,et al.  Quantile regression for the single-index coefficient model , 2017 .

[13]  M. Kolar,et al.  Uniform Inference for High-dimensional Quantile Regression: Linear Functionals and Regression Rank Scores , 2017, 1702.06209.

[14]  Runze Li,et al.  Local composite quantile regression smoothing: an efficient and safe alternative to local polynomial regression , 2010 .

[15]  Rong Jiang,et al.  Single-index composite quantile regression for massive data , 2020, J. Multivar. Anal..

[16]  Cun-Hui Zhang,et al.  Confidence Intervals for Low-Dimensional Parameters With High-Dimensional Data , 2011 .

[17]  W. Härdle,et al.  Semi-parametric estimation of partially linear single-index models , 2006 .

[18]  Martin J. Wainwright,et al.  Fast mixing of Metropolized Hamiltonian Monte Carlo: Benefits of multi-step gradients , 2019, J. Mach. Learn. Res..

[19]  Yaniv Plan,et al.  The Generalized Lasso With Non-Linear Observations , 2015, IEEE Transactions on Information Theory.

[20]  Guang Cheng,et al.  Simultaneous Inference for High-Dimensional Linear Models , 2016, 1603.01295.

[21]  Yan Yu,et al.  Single-index quantile regression , 2010, J. Multivar. Anal..

[22]  A. Belloni,et al.  L1-Penalized Quantile Regression in High Dimensional Sparse Models , 2009, 0904.2931.

[23]  Yuankun Zhang.  Ultra-High Dimensional Single-Index Quantile Regression , 2020 .

[24]  Yan Yu,et al.  Estimation and variable selection for quantile partially linear single-index models , 2017, J. Multivar. Anal..

[25]  Cun-Hui Zhang,et al.  Confidence intervals for low dimensional parameters in high dimensional linear models , 2011, 1110.2563.

[26]  Ying Zhu,et al.  High Dimensional Inference in Partially Linear Models , 2019, AISTATS.

[27]  Xi Chen,et al.  Distributed High-dimensional Regression Under a Quantile Loss Function , 2019, J. Mach. Learn. Res..

[28]  Rong Jiang,et al.  Weighted composite quantile regression for single-index models , 2016, J. Multivar. Anal..

[29]  Peter Radchenko,et al.  High dimensional single index models , 2015, J. Multivar. Anal..

[30]  Qianqian Zhu,et al.  Estimation of linear composite quantile regression using EM algorithm , 2016 .

[31]  Eliana Christou,et al.  Single index quantile regression for heteroscedastic data , 2016, J. Multivar. Anal..

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

[33]  Claudia Czado,et al.  D-vine copula based quantile regression , 2015, Comput. Stat. Data Anal..

[34]  Peng Zhao,et al.  On Model Selection Consistency of Lasso , 2006, J. Mach. Learn. Res..

[35]  Adel Javanmard,et al.  Confidence intervals and hypothesis testing for high-dimensional regression , 2013, J. Mach. Learn. Res..

[36]  Linjun Tang,et al.  Weighted local linear CQR for varying-coefficient models with missing covariates , 2015 .

[37]  W. Härdle,et al.  Single-Index-Based CoVaR With Very High-Dimensional Covariates , 2018 .

[38]  Linglong Kong,et al.  Advanced algorithms for penalized quantile and composite quantile regression , 2021, Comput. Stat..

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

[40]  Aurélien Garivier,et al.  On the Complexity of Best-Arm Identification in Multi-Armed Bandit Models , 2014, J. Mach. Learn. Res..