In many real-life object allocation problems, objects' priorities involve indifference classes. The common practice to deal with ties is to randomly break them to obtain strict priorities. From the ex-ante point of view, random tie breaking induces random priorities. In this setting, Dr. Afacan will introduce a fairness notion generalizing the usual stability of Gale and Shapley (1962) to the random priorities setting. Then, he will check whether the well-adopted agent-optimal stable mechanism satisfies that notion along with other desirable properties. Dr. Afacan will next contruct a mechanism mainly built on the Probabilistic Serial Mechanism of Bogomolnaia and Moulin (2001). He will define that new rule in a way that it will dominate the agent-optimal stable mechanism on the efficiency and fairness grounds. Lastly, he will characterize the introduced new mechanism.
Call for proposal
See other projects for this call