- This event has passed.
Special LSE PhD student session: Goreti Faria & James Nguyen
4 May 2016, 5:30 pm – 7:00 pm
Goreti Faria: “A Preference for Late Resolution of Uncertainty”
Kreps and Porteus (1978) (KP) offer an axiomatic approach to dynamic decision problems that allows us to explicitly model preferences regarding how a decision is made. In particular, their model makes it possible to explicitly model preferences for the timing of the resolution of uncertainty. In the orthodox framework (vNM) uncertainties resolving at different times are indistinguishable from one another. I present an example where the agent has a preference for early resolution of uncertainty, and I solve it using KP’s model. I then argue that the only way the orthodox model can accommodate such a preference is via an unsatisfactory individuation strategy, and I argue that the advantages of KP’s modelling of the situation surpass its disadvantages.
James Nguyen: “Moving Beyond Arrow’s Theorem: Social Choice and Theory Choice”
Okasha (2011) suggests that the problem of theory choice in science — how to choose between multiple competing theories, models, or scientific alternatives — can be construed as a social choice problem. His gambit is to identify theoretical virtues (accuracy, simplicity, scope …) as providing preference rankings over the alternatives, and theory choice becomes a matter of aggregating these rankings. But if the Arrovian conditions apply, then rational theory choice is impossible. In this talk I investigate two possible escape routes. Firstly, as suggested by Morreau (2015) some scientific virtues are `rigid’, in the sense that they cannot but provide the preference ranking they provide. I prove a possibility result over the resulting restricted domain, but show that it comes at a cost. Secondly, I investigate aggregating to a non-binary choice function, as opposed to a preference ranking. I argue that it is unclear what sort of independence (of irrelevant alternatives) condition applies in this context, and offer a novel way of proving Sen’s (1993) impossibility result that demonstrates the role played by his `Independent Decisiveness’ condition.