Affordable Access

Empirical risk minimization is consistent with the mean absolute percentage error

Authors
  • De Myttenaere, Arnaud
  • Grand, Bénédicte Le
  • Rossi, Fabrice
Type
Preprint
Publication Date
Sep 08, 2015
Submission Date
Sep 08, 2015
Identifiers
arXiv ID: 1509.02357
Source
arXiv
License
Yellow
External links

Abstract

We study in this paper the consequences of using the Mean Absolute Percentage Error (MAPE) as a measure of quality for regression models. We show that finding the best model under the MAPE is equivalent to doing weighted Mean Absolute Error (MAE) regression. We also show that, under some asumptions, universal consistency of Empirical Risk Minimization remains possible using the MAPE.

Report this publication

Statistics

Seen <100 times