Kybernetika 53 no. 3, 480-492, 2017

Directional quantile regression in R

Pavel Boček and Miroslav ŠimanDOI: 10.14736/kyb-2017-3-0480

Abstract:

Recently, the eminently popular standard quantile regression has been generalized to the multiple-output regression setup by means of directional regression quantiles in two rather interrelated ways. Unfortunately, they lead to complicated optimization problems involving parametric programming, and this may be the main obstacle standing in the way of their wide dissemination. The presented R package modQR is intended to address this issue. It originates as a quite faithful translation of the authors' moQuantile toolbox for Octave and MATLAB, and provides all the necessary computational support for both the directional multiple-output quantile regression methods to the wide statistical public. The article offers a concise summary of the statistical theory behind modQR, overviews the package in brief, points out its departures from moQuantile, comments on its use and performance, and demonstrates its application.

Keywords:

multivariate quantile, regression quantile, halfspace depth, regression depth, depth contour

Classification:

62-04, 65C60, 62H05, 62J99

paper.pdf

References:

  1. P. Boček and M. Šiman: modQR: Multiple-Output Directional Quantile Regression. R package version 0.1.0, 2015.   CrossRef
  2. P. Boček and M. Šiman: Directional quantile regression in Octave and MATLAB. Kybernetika 52 (2016), 28-51.   DOI:10.14736/kyb-2016-1-0028
  3. B. Chakraborty: On multivariate quantile regression. J. Statist. Planning Inference 110 (2003), 109-132.   DOI:10.1016/s0378-3758(01)00277-4
  4. I. Charlier, D. Paindaveine and J. Saracco: Multiple-output regression through optimal quantization. ECARES Working Paper 2016-18.   CrossRef
  5. P. Chaudhury: On a geometric notion of quantiles for multivariate data. J. Amer. Stat. Assoc. 91 (1996), 862-872.   DOI:10.2307/2291681
  6. Y. Cheng and J. G. De Gooijer: On the $u$th geometric conditional quantile. J. Statist. Planning Inference 137 (2007), 1914-1930.   DOI:10.1016/j.jspi.2006.02.014
  7. Š. Došlá: Conditions for bimodality and multimodality of a mixture of two unimodal densities. Kybernetika 45 (2009) 279-292.   CrossRef
  8. M. Hallin, Z. Lu, D. Paindaveine and M. Šiman: Local bilinear multiple-output quantile/depth regression. Bernoulli 21 (2015), 1435-1466.   DOI:10.3150/14-bej610
  9. M. Hallin, D. Paindaveine and M. Šiman: Multivariate quantiles and multiple-output regression quantiles: From ${L}_1$ optimization to halfspace depth. Ann. Statist. 38 (2010), 635-669.   DOI:10.1214/09-aos723
  10. M. Hallin, D. Paindaveine and M. Šiman: Rejoinder. Ann. Statist. 38 (2010), 694-703.   DOI:10.1214/09-aos723rej
  11. R. Koenker: Quantile Regression. Cambridge University Press, New York 2005.   DOI:10.1017/cbo9780511754098
  12. R. Koenker and G. J. Bassett: Regression quantiles. Econometrica 46 (1978), 33-50.   DOI:10.2307/1913643
  13. V. Koltchinskii: ${M}$-estimation, convexity and quantiles. Ann. Statist. 25 (1997), 435-477.   DOI:10.1214/aos/1031833659
  14. L. Kong and I. Mizera: Quantile tomography: Using quantiles with multivariate data. Statistica Sinica 22 (2012), 1589-1610.   DOI:10.5705/ss.2010.224
  15. I. W. McKeague, S. López-Pintado, M. Hallin and M. Šiman: Analyzing growth trajectories. J. Developmental Origins of Health and Disease 2 (2011), 322-329.   DOI:10.1017/s2040174411000572
  16. D. Paindaveine and M. Šiman: On directional multiple-output quantile regression. J. Multivariate Anal. 102 (2011), 193-212.   DOI:10.1016/j.jmva.2010.08.004
  17. D. Paindaveine and M. Šiman: Computing multiple-output regression quantile regions. Comput. Statist. Data Anal. 56 (2012), 840-853.   DOI:10.1016/j.csda.2010.11.014
  18. D. Paindaveine and M. Šiman: Computing multiple-output regression quantile regions from projection quantiles. Comput. Statist. 27 (2012), 29-49.   DOI:10.1007/s00180-011-0231-y
  19. M. Šiman: On exact computation of some statistics based on projection pursuit in a general regression context. Commun. Statist. - Simulation and Computation 40 (2011), 948-956.   DOI:10.1080/03610918.2011.560730
  20. M. Šiman: Precision index in the multivariate context. Commun. Statist. - Theory and Methods 43 (2014), 377-387.   DOI:10.1080/03610926.2012.661509