Electronic Journal of Statistics

Variable selection of varying coefficient models in quantile regression

Hohsuk Noh, Kwanghun Chung, and Ingrid Van Keilegom

Full-text: Open access

Abstract

Varying coefficient (VC) models are commonly used to study dynamic patterns in many scientific areas. In particular, VC models in quantile regression are known to provide a more complete description of the response distribution than in mean regression. In this paper, we develop a variable selection method for VC models in quantile regression using a shrinkage idea. The proposed method is based on the basis expansion of each varying coefficient and the regularization penalty on the Euclidean norm of the corresponding coefficient vector. We show that our estimator is obtained as an optimal solution to the second order cone programming (SOCP) problem and that the proposed procedure has consistency in variable selection under suitable conditions. Further, we show that the estimated relevant coefficients converge to the true functions at the univariate optimal rate. Finally, the method is illustrated with numerical simulations including the analysis of forced expiratory volume (FEV) data.

Article information

Source
Electron. J. Statist., Volume 6 (2012), 1220-1238.

Dates
First available in Project Euclid: 9 July 2012

Permanent link to this document
https://projecteuclid.org/euclid.ejs/1341842803

Digital Object Identifier
doi:10.1214/12-EJS709

Mathematical Reviews number (MathSciNet)
MR2988445

Zentralblatt MATH identifier
1295.62072

Subjects
Primary: 62G35: Robustness
Secondary: 62J07: Ridge regression; shrinkage estimators

Keywords
Basis approximation consistency in variable selection second order cone programming shrinkage estimator

Citation

Noh, Hohsuk; Chung, Kwanghun; Van Keilegom, Ingrid. Variable selection of varying coefficient models in quantile regression. Electron. J. Statist. 6 (2012), 1220--1238. doi:10.1214/12-EJS709. https://projecteuclid.org/euclid.ejs/1341842803


Export citation

References

  • [1] Alizadeh, F. and D. Goldfarb (2003). Second-order cone programming., Mathematical Programming B95, 3–51.
  • [2] Cai, Z. and X. Xu (2008). Nonparametric quantile estimations for dynamic smooth coefficient models., Journal of the American Statistical Association 103, 1595–1608.
  • [3] de Boor, C. (1978)., A Practical Guide to Splines. Springer-Verlag.
  • [4] Fan, J. and R. Li (2001). Variable selection via nonconcave penalized likelihood and its oracle properties., Journal of the American Statistical Association 99, 710–723.
  • [5] Grant, M. and S. Boyd (2011, April). CVX: Matlab software for disciplined convex programming, version 1.21., http://cvxr.com/cvx.
  • [6] Honda, T. (2004). Quantile regression in varying coefficient models., Journal of Statistical Planning and Inference 121, 113–125.
  • [7] Huang, J. Z. and L. Yang (2004). Identification of non-linear additive autoregressive models., Journal of the Royal Statistical Society B66, 463–477.
  • [8] Kai, B., R. Li, and H. Zou (2011). New efficient estimation and variable selection methods for semiparametric varying-coefficient partially linear models., Annals of Statistics 39, 305–332.
  • [9] Kim, M. O. (2007). Quantile regression with varying coefficients., Annals of Statistics 35, 92–108.
  • [10] Koenker, R., P. NG, and S. Portnoy (1994). Quantile smoothing splines., Biometrika 81, 673–680.
  • [11] Lee, E. R. and H. Noh (2012). Model selection via Bayesian information critetion for quantile regression models. Technical report, Université catholique de, Louvain.
  • [12] Li, Y. and J. Zhu (2008). L1-norm quantile regression., Journal of Computational and Graphical Statistics 17, 163–185.
  • [13] Lobo, M., L. Vandenberghe, S. Boyd, and H. Lebret (1998). Applications of second-order cone programming., Linear Algebra and its Applications 284, 193–228.
  • [14] Noh, H. and B. Park (2010). Sparse varying coefficient models for longitudinal data., Statistica Sinica 20, 1183–1202.
  • [15] Rosner, B. (2000)., Fundamentals of Biostatistics. Duxbury.
  • [16] Ruppert, D. (2002). Selecting the number of knots for penalized splines., Journal of Computational and Graphical Statistics 11, 735–757.
  • [17] Schwarz, G. (1978). Estimating the dimension of a model., Annals of Statistics 6, 461–464.
  • [18] Stone, C. J. (1982). Optimal global rates of convergence for nonparametric regression., Annals of Statistics 10, 1348–1360.
  • [19] Storlie, C. B., H. D. Bondell, B. J. Reich, and H. H. Zhang (2011). Surface estimation, variable selection, and the nonparametric oracle property., Statistica Sinica 21, 679–705.
  • [20] Tang, Y., H. J. Wang, Z. Zhu, and X. Song (2012). A unified variable selection approach for varying coefficient models., Statistica Sinica 22, 601–628.
  • [21] Wang, H. and Y. Xia (2009). Shrinkage estimation of the varying coefficient model., Journal of the American Statistical Association 104, 747–757.
  • [22] Wang, H. J., Z. Zhu, and J. Zhou (2009). Quantile regression in partially linear varying coefficient models., Annals of Statistics 37, 3841–3866.
  • [23] Wang, L., H. Li, and J. Z. Huang (2008). Variable selection in nonparametric varying-coefficient models for analysis of repeated measurements., Journal of the American Statistical Association 103, 1556–1569.
  • [24] Xue, L. (2009). Variable selection in additive models., Statistica Sinica 19, 1281–1296.
  • [25] Yuan, M. and Y. Lin (2006). Model selection and estimation in regression with grouped variables., Journal of the Royal Statistical Society B 68, 49–67.
  • [26] Zou, H. (2006). The adaptive lasso and its oracle properties., Journal of the American Statistical Association 101, 1418–1429.
  • [27] Zou, H. and R. Li (2008). One-step sparse estimates in nonconcave penalized likelihood models., Annals of Statistics 36, 1509–1533.