On convergence rates of convex regression in multiple dimensions

Document Type

Article

Publication Date

1-1-2014

Abstract

We consider a least squares estimator for estimating a convex function f*: [0,1]d → ℝ with bounded sub-gradients. A rate at which the sum of squared differences between the estimator and the true function f* converges to zero is computed. This work sheds light on computing the convergence rate of the multidimensional convex regression estimator. © 2014 INFORMS.

Publication Title

INFORMS Journal on Computing

First Page Number

616

Last Page Number

628

DOI

10.1287/ijoc.2013.0587

This document is currently not available here.

Share

COinS