Nonparametric Frontier Estimation by Linear Programming - Inria - Institut national de recherche en sciences et technologies du numérique Access content directly
Journal Articles Automation and Remote Control / Avtomatika i Telemekhanika Year : 2004

Nonparametric Frontier Estimation by Linear Programming

Abstract

A new method for estimating the frontier of a set of points (or a support, in other words) is proposed. The estimates are defined as kernel functions covering all the points and whose associated support is of smallest surface. They are written as linear combinations of kernel functions applied to the points of the sample. The weights of the linear combination are then computed by solving a linear programming problem. In the general case, the solution of the optimization problem is sparse, that is, only a few coefficients are non zero. The corresponding points play the role of support vectors in the statistical learning theory. The L 1-norm for the error of estimation is shown to be almost surely converging to zero, and the rate of convergence is provided.

Keywords

Dates and versions

inria-00548244 , version 1 (20-12-2010)

Identifiers

Cite

Guillaume Bouchard, Stéphane Girard, Anatoli B. Juditsky, Alexander Nazin. Nonparametric Frontier Estimation by Linear Programming. Automation and Remote Control / Avtomatika i Telemekhanika, 2004, 65 (1), pp.58--64. ⟨10.1023/B:AURC.0000011690.13220.58⟩. ⟨inria-00548244⟩
61 View
0 Download

Altmetric

Share

Gmail Facebook Twitter LinkedIn More