Dynamic pricing and learning with competition : Insights from the Dynamic Pricing Challenge at the 2017 INFORMS RM & Pricing Conference

Abstract

This paper presents the results of the Dynamic Pricing Challenge, held on the occasion of the 17th INFORMS Revenue Management and Pricing Section Conference on June 29-30, 2017, at the Centrum Wiskunde & Informatica, Amsterdam, The Netherlands. The participants of this pricing challenge submitted a wide variety of pricing and learning algorithms of which the numerical performance in a simulated market environment with competition was analyzed. This allows to consider market dynamics that are not analytically tractable or can not be empirically analyzed due to practical complications. Our findings implicate that the relative performance of algorithms varies substantially across oligopoly and duopoly markets and across different market dynamics, which confirms the inherit complexity of pricing and learning in the presence of competition. Most notably, none of the considered algorithms is able to consistently outperform the market. Furthermore, wefind that algorithms vary substantially in terms of robustness, that ignoring competition is increasingly harmful as competition is more fierce, and that too much exploration hurts performance significantly.

More information

Identifiers

ORCID for Christine Currie: orcid.org/0000-0002-7016-3652

ORCID for Antonio Martinez Sykora: orcid.org/0000-0002-2435-3113

ORCID for Simos Zachariades: orcid.org/0000-0002-1916-0018

Catalogue record

Export record

Altmetrics