We propose a new proxy bidding mechanism to allocate courses to students given students' reported preferences. Our mechanism is motivated by a specific strategic downgrading manipulation observed in the course allocation mechanism currently used at Harvard Business School (HBS). The proxy bidding mechanism simplifes students' decisions by directly incorporating downgrading into the mechanism. Our proxy bidding mechanism is Pareto efficient with respect to lexicographic preferences and can be extended to allow for a more expressive preference language which allows complementarity, substitutability, and indifference. Simulations suggest that the proxy bidding mechanism is robust to the manipulations observed in the HBS mechanism and may yield welfare improvements. © 2010 Springer-Verlag.
CITATION STYLE
Kominers, S. D., Ruberry, M., & Ullman, J. (2010). Course allocation by proxy auction. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6484 LNCS, pp. 551–558). https://doi.org/10.1007/978-3-642-17572-5_49
Mendeley helps you to discover research relevant for your work.