Risk bounds for quantile trend filtering

Oscar Hernan Madrid Padilla, Sabyasachi Chatterjee

Research output: Contribution to journalArticlepeer-review


We study quantile trend filtering, a recently proposed method for nonparametric quantile regression, with the goal of generalizing existing risk bounds for the usual trend-filtering estimators that perform mean regression. We study both the penalized and the constrained versions, of order r ⩾ 1, of univariate quantile trend filtering. Our results show that both the constrained and the penalized versions of order r ⩾ 1 attain the minimax rate up to logarithmic factors, when the (r-1) th discrete derivative of the true vector of quantiles belongs to the class of bounded-variation signals. Moreover, we show that if the true vector of quantiles is a discrete spline with a few polynomial pieces, then both versions attain a near-parametric rate of convergence. Corresponding results for the usual trend-filtering estimators are known to hold only when the errors are sub-Gaussian. In contrast, our risk bounds are shown to hold under minimal assumptions on the error variables. In particular, no moment assumptions are needed and our results hold under heavy-tailed errors. Our proof techniques are general, and thus can potentially be used to study other nonparametric quantile regression methods. To illustrate this generality, we employ our proof techniques to obtain new results for multivariate quantile total-variation denoising and high-dimensional quantile linear regression.

Original languageEnglish (US)
Pages (from-to)751-768
Number of pages18
Issue number3
StatePublished - Sep 1 2022


  • Adaptation
  • Huber loss
  • Local Rademacher width
  • Quantile regression
  • Trend filtering

ASJC Scopus subject areas

  • Statistics and Probability
  • General Mathematics
  • Agricultural and Biological Sciences (miscellaneous)
  • General Agricultural and Biological Sciences
  • Statistics, Probability and Uncertainty
  • Applied Mathematics


Dive into the research topics of 'Risk bounds for quantile trend filtering'. Together they form a unique fingerprint.

Cite this