Informs award for CWI Stochastics researchers

Researchers Bert Zwart (ST) and former PhD student Arnoud den Boer have been awarded the INFORMS Revenue Management and Pricing Section Award at the INFORMS Annual Meeting 2016 in Nashville.

Publication date
21 Nov 2016

Researchers Bert Zwart (ST) and former PhD student Arnoud den Boer have been awarded the INFORMS Revenue Management and Pricing Section Award at the INFORMS Annual Meeting 2016 in Nashville. The Institute for Operations Research and the Management Sciences (INFORMS) is the largest society in the world for professionals in the field of operations research (O.R.), management science, and analytics. The prize has been awarded for the best contribution to the science of pricing and revenue management published in English.

The researchers of CWI received the prize for the following publications:

Den Boer, A., Zwart, B. (2014). Simultaneously Learning and Optimizing Using
Controlled Variance Pricing. Management Science 60, 770 – 783.

Den Boer, A., Zwart, B. (2015). Dynamic pricing and learning with finite
inventories. Operations Research 63, 965 – 979.

Den Boer, A., Zwart, B. (2014). Mean square convergence rates for maximum
quasi-likelihood estimators. Stochastic Systems 4, 375 – 403.

Informs regards this as a strongly connected group of papers where iterative pricing strategies are considered using parametric demand models, such as linear models or logit models.

The first paper partially resolved an issue that had been open for more than 35 years, showing that a greedy pricing policy may be useful in the short run, but leads to suboptimal profit in the long run. This paper then proposes a new pricing policy called controlled variance pricing that solves this problem.
In the second paper, it is shown that a greedy policy works in case inventories are relatively small: the optimal price fluctuates with the remaining inventory, and therefore it is possible to learn price elasticities of consumers.

These are important operational insights and are based on new results in mathematical statistics, which are developed in the third paper.

The results have also found their way in practice: a pilot in an online wine company
(in collaboration with Arenoe Marketing Intelligence) has shown that the methods work well.