MPI TAX
Contests & Conflict | 06.07.2022 | 16:00
Optimal Contest Design with Tree Architecture

Referent: Qian Jiao

Date & Time: Los Angeles (7 am), Cincinnati (10 am), Bath (3 pm), Munich (4 pm), Beijing (10 pm), Singapore (10 pm), Sydney (July 7, 12 am)

Qian Jiao will present the paper titled "Optimal Contest Design with Tree Architecture" (joint work with Zhonghong Kuang, Yiran Liu, and Yang Yu).

Abstract of the Paper:
This paper investigates the optimal (effort-maximizing) design of tree contests. The contest organizer is allowed to design the contest architecture and its associated prize structure. We first propose the pivotal match principle to rationalize the optimality of the winner-take-all prize allocation rule and then fully characterize the optimal contest architecture. We find that when the contest technology is relatively discriminatory, the simultaneous contest is optimal; in addition, when the contest technology is sufficiently noisy, the contest with a balanced binary tree architecture is optimal, in which case each contestant experiences at most one bye. Finally, when the contest technology is moderately noisy, each embedded match should include a prime number of participants, the optimal architecture can be solved by a dynamic programming algorithm. 

Chair: Tracy Liu, Co-Chair: Qiang Fu and Lionel Page

Ansprechpartner

Event Team

Max-Planck-Institut für Steuerrecht und Öffentliche Finanzen

Marstallplatz 1
80539 München

Telefon: +49-89-24246-5255
Fax: +49-89-24246-5299

E-Mail: contests@tax.mpg.de