Seminar. Duncan James. “Risky Curves–New Results and Future Directions”

To join the seminar, sign up here.

Duncan James (Fordham University) will speak about his book “Risky Curves: On the empirical failure of expected utility,” co-authored with Daniel Friedman, Mark Isaac, and Shyam Sunder.

Risky Curves — New Results and Future Directions

The book summarizes 8 decades of empirical work in expected utility theory and its generalizations. It comes to the sobering conclusion that its “power to predict out-of-sample is in the poor-to-non-existent range.” Estimated parameters vary wildly across domains (like individual choice vs. aggregate behaviour), contexts, and even within the same individual over short time spans. In addition to revisiting the findings of the book, the seminar also includes a deeper look into new results extending points raised in the book.

We invited the authors of Risky Curves because their findings dovetail nicely with our own work in Ergodicity Economics, which provides a physical, dynamic, interpretation of the “curves” of expected utility theory (utility functions). EE suggests that apparent utility functions are much more reflections of the dynamic context of the decision maker than of idiosyncratic preferences. When interpreted idiosyncratically as in expected-utility theory, one would expect inconsistencies arising from changing contexts in which decision makers operate. These contexts are typically ignored in expected-utility theory.

To receive invitations to future seminars in the ergodicity economics series, sign up to our mailing list.