Honors Theses
Date of Award
Spring 5-10-2025
Document Type
Undergraduate Thesis
Department
Economics
First Advisor
Mark Van Boening
Second Advisor
John Conlon
Third Advisor
Josh Hendrickson
Relational Format
Dissertation/Thesis
Abstract
I present the use of best response (BR) and imitation (IM) learning rules in experimental Cournot duopoly markets under asymmetric information. In treatment BR, one subject (called BR1) receives, after every round, information about their previous quantity choices, the previous market prices, their previous profits, the previous quantities of the the other firm in the market, and the previous Cournot best response to the quantity of the other firm. In treatment IM, one subject (called IM1) receives, after every round, information about the prior quantities of and profits of both firms in the market. The BR1 and IM1 subjects are paired with BR2 and IM2 subjects, respectively. After every round, subjects BR2 and IM2 are only able to view their own quantity decisions, the market prices, and their profits from every previous round, and not the profits of other subjects or the best response to the choices of the other subject. I find evidence that the asymmetric availability of information about the Cournot game results in a leader-follower dynamic. In treatment BR, the BR2 subject typically emerges as the leader, while in treatment IM, the IM1 subject typically emerges as the leader.
Recommended Citation
Pentes, Marshall H., "Learning Rules under Asymmetric Information: A Cournot Market Experiment" (2025). Honors Theses. 3264.
https://egrove.olemiss.edu/hon_thesis/3264
Description of spreadsheets and regression code
BR Database.xlsx (162 kB)
Full dataset for BR sessions
IM Database.xlsx (281 kB)
Full dataset for IM sessions
BR EWA.xlsx (29 kB)
BR1 dataset adjusted for EWA model
IM EWA.xlsx (46 kB)
IM1 dataset adjusted for EWA model
BR Huck model.xlsx (148 kB)
BR1 dataset adjusted for Huck model
IM Huck model.xlsx (190 kB)
IM1 dataset adjusted for Huck model
Creative Commons License

This work is licensed under a Creative Commons Attribution 4.0 International License.