Statistics and Computing

, Volume 19, Issue 3, pp 271–281 | Cite as

Automatic Bayesian quantile regression curve fitting

  • Colin ChenEmail author
  • Keming Yu


Quantile regression, including median regression, as a more completed statistical model than mean regression, is now well known with its wide spread applications. Bayesian inference on quantile regression or Bayesian quantile regression has attracted much interest recently. Most of the existing researches in Bayesian quantile regression focus on parametric quantile regression, though there are discussions on different ways of modeling the model error by a parametric distribution named asymmetric Laplace distribution or by a nonparametric alternative named scale mixture asymmetric Laplace distribution. This paper discusses Bayesian inference for nonparametric quantile regression. This general approach fits quantile regression curves using piecewise polynomial functions with an unknown number of knots at unknown locations, all treated as parameters to be inferred through reversible jump Markov chain Monte Carlo (RJMCMC) of Green (Biometrika 82:711–732, 1995). Instead of drawing samples from the posterior, we use regression quantiles to create Markov chains for the estimation of the quantile curves. We also use approximate Bayesian factor in the inference. This method extends the work in automatic Bayesian mean curve fitting to quantile regression. Numerical results show that this Bayesian quantile smoothing technique is competitive with quantile regression/smoothing splines of He and Ng (Comput. Stat. 14:315–337, 1999) and P-splines (penalized splines) of Eilers and de Menezes (Bioinformatics 21(7):1146–1153, 2005).


Asymmetric Laplace distribution Piecewise polynomials Quantile regression Reversible jump Markov chain Monte Carlo Splines 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. Chaudhuri, P.: Nonparametric quantile regression. Ann. Stat. 19, 760–777 (1991) zbMATHCrossRefMathSciNetGoogle Scholar
  2. Chen, C.: A finite smoothing algorithm for quantile regression. J. Comput. Graph. Stat. 16, 136–164 (2007) CrossRefGoogle Scholar
  3. Chu, C.K., Glad, I.K., Godtliebsen, F., Marron, J.S.: Edge-preserving smoothers for image processing (with discussion). J. Am. Stat. Assoc. 93, 526–541 (1998) zbMATHCrossRefMathSciNetGoogle Scholar
  4. Denison, D., Mallick, B., Smith, A.: Automatic Bayesian curve fitting. J. R. Stat. Soc. Ser. B 60, 333–350 (1998) zbMATHCrossRefMathSciNetGoogle Scholar
  5. DiMatteo, I., Genovese, C.R., Kass, R.E.: Bayesian curve fitting with free-knot splines. Biometrika 88, 1055–1073 (2001) zbMATHCrossRefMathSciNetGoogle Scholar
  6. Eilers, P., de Menezes, R.: Quantile smoothing of array CGH data. Bioinformatics 21(7), 1146–1153 (2005) CrossRefGoogle Scholar
  7. Green, P.J.: Reversible jump Markov chain Monte Carlo computation and Bayesian model determination. Biometrika 82, 711–732 (1995) zbMATHCrossRefMathSciNetGoogle Scholar
  8. Hansen, M.H., Kooperberg, C.: Spline adaptation in extended linear models (with discussion). Stat. Sci. 17, 2–51 (2002) zbMATHCrossRefMathSciNetGoogle Scholar
  9. He, X., Ng, P.: COBS: Qualitatively constrained smoothing via linear program. Comput. Stat. 14, 315–337 (1999) zbMATHCrossRefGoogle Scholar
  10. Kass, R.E., Wallstrom, G.L.: Comment on: Spline adaptation in extended linear models by Mark H. Hansen and Charles Kooperberg. Stat. Sci. 17, 2–51 (2002) CrossRefGoogle Scholar
  11. Koenker, R.: Quantile Regression. Cambridge University Press, Cambridge (2005) zbMATHGoogle Scholar
  12. Koenker, R., Bassett, J.G.: Regression quantiles. Econometrica 46, 33–50 (1978) zbMATHCrossRefMathSciNetGoogle Scholar
  13. Koenker, R., Ng, P., Portnoy, S.: Quantiles smoothing splines. Biometrika 81, 673–680 (1994) zbMATHCrossRefMathSciNetGoogle Scholar
  14. Leslie, D.S., Kohn, R., Nott, D.J.: A general approach to heteroscedastic linear regression. Stat. Comput. 17, 131–146 (2007) CrossRefMathSciNetGoogle Scholar
  15. Smith, M., Kohn, R.: Nonparametric regression using Bayesian variable selection. J. Econom. 75, 317–343 (1996) zbMATHCrossRefGoogle Scholar
  16. Stone, C.J.: Nonparametric M-regression with free knots. J. Stat. Plan. Inference 130, 183–206 (2005) zbMATHCrossRefGoogle Scholar
  17. Silverman, B.W.: Some aspects of the spline smoothing approach to non-parametric regression curve fitting. J. R. Stat. Soc. Ser. B 47, 1–52 (1985) zbMATHGoogle Scholar
  18. Yu, K.: Reversible jump MCMC approach quantile regression. Comput. Stat. Data Anal. 40(2), 303–315 (2002) zbMATHCrossRefGoogle Scholar
  19. Yu, K., Jones, M.C.: Local linear quantile regression. J. Am. Stat. Assoc. 93, 228–238 (1998) zbMATHCrossRefMathSciNetGoogle Scholar
  20. Yu, K., Moyeed, R.A.: Bayesian quantile regression. Stat. Probab. Lett. 54, 437–447 (2001) zbMATHCrossRefMathSciNetGoogle Scholar
  21. Zhou, S., Shen, X.: Spatially adaptive regression splines and accurate knot selection schemes. J. Am. Stat. Assoc. 96, 247–259 (2001) zbMATHCrossRefMathSciNetGoogle Scholar

Copyright information

© Springer Science+Business Media, LLC 2008

Authors and Affiliations

  1. 1.SAS Institute Inc.CaryUSA
  2. 2.Department of Mathematical SciencesBrunel UniversityUxbridgeUK

Personalised recommendations