An Exact Pointwise Characterization for Total Variation Denoising in Quantile Regression

TL;DR

This paper provides an exact pointwise characterization of the quantile total variation denoising estimator, extending recent mean regression results and revealing structural properties and risk bounds under heavy-tailed noise.

Contribution

It derives a novel minmax/maxmin representation for the quantile TVD estimator, demonstrating non-crossing across quantiles and enabling refined pointwise analysis.

Findings

The set of admissible fitted values forms a compact interval characterized by local order statistics.

Quantile TVD solutions are closed under coordinatewise maximum and minimum, ensuring extremal solutions.

The representation allows for near-optimal pointwise risk bounds under heavy-tailed noise.

Abstract

Total variation denoising (TVD) is a classical method for denoising and curve fitting, yet an explicit pointwise description of its fitted values has only recently been established in the mean regression setting by arXiv:2410.03041v4. This raises the question of whether a similar representation holds for quantile regression. We answer this question affirmatively by deriving an exact minmax/maxmin representation for the quantile TVD estimator, providing a complete pointwise characterization of its solution set. Given that the quantile TVD estimator is generally non-unique, the existence of such a representation is perhaps surprising. We show that the set of admissible fitted values at any location forms a compact interval, whose endpoints are characterized exactly by minmax/maxmin functionals of local order statistics over nested intervals. We next develop several structural properties of the quantile TVD solution set. First, the solution set is closed under coordinatewise maximum and minimum, guaranteeing the existence of extremal elements -- upper and lower envelope solutions. Second, this reveals that quantile TVD is intrinsically non-crossing across quantile levels when a common tuning parameter is used. We prove this is driven by submodularity of the total variation penalty, and show that any penalized quantile regression estimator with a submodular penalty enjoys this property. From an estimation error perspective, our representation enables a refined pointwise analysis via a transparent local bias-variance decomposition, facilitating new pointwise risk bounds and near-optimal rates for locally Holder smooth functions. Our results hold under heavy-tailed noise (e.g., Cauchy) and substantially extend existing guarantees beyond locally constant signals. Altogether, these results advance the theory of quantile TV regression via exact pointwise min-max representations.