Least Squares and Shrinkage Estimation under Bimonotonicity Constraints

TL;DR

This paper introduces algorithms for minimizing smooth functions under order constraints, focusing on bimonotone matrices, with applications in regression, shrinkage estimation, and image denoising, supported by numerical examples.

Contribution

It develops active set algorithms for bimonotonicity constraints, enabling efficient estimation in regression and denoising tasks with order restrictions.

Findings

Algorithms effectively handle bimonotonicity constraints.

Applications include regression and image denoising.

Numerical examples demonstrate practical performance.

Abstract

In this paper we describe active set type algorithms for minimization of a smooth function under general order constraints, an important case being functions on the set of bimonotone r-by-s matrices. These algorithms can be used, for instance, to estimate a bimonotone regression function via least squares or (a smooth approximation of) least absolute deviations. Another application is shrinkage estimation in image denoising or, more generally, regression problems with two ordinal factors after representing the data in a suitable basis which is indexed by pairs (i,j) in {1,...,r}x{1,...,s}. Various numerical examples illustrate our methods.