Roman Frydman is Professor of Economics at New York University and Chair of the INET Program on Imperfect Knowledge Economics. Frydman was one of the early critics of the Rational Expectations Hypothesis, arguing in a 1982 article in the American Economic Review that REH models do not provide an adequate representation of rational decision-making in real-world markets.
Over the last two decades, his work with Michael Goldberg traced the epistemological and empirical difficulties of existing macroeconomic and finance models to their core premise that economists can represent how outcomes unfold over time with a stochastic process. They presented this research in Imperfect Knowledge Economics (Princeton University Press, 2007) and Beyond Mechanical Markets PUP, 2011). Recently, Frydman co-edited (with Edmund S. Phelps), Rethinking Expectations: The Way Forward for Macroeconomics (PUP, 2013), which examines alternative approaches that aim to shape the research agenda in macroeconomics and policy analysis.
Building on this research, Frydman, in collaboration with Søren Johansen, Anders Rahbek and Morten Tabor, has developed an approach to consistent model building under Knightian uncertainty. By recognizing the inherent limits to what we can know about change, their novel mathematical framework, which they call Knightian Uncertainty Expectations (KUE), reveals a new way to understand how both fundamental and psychological considerations drive rational behavior and market outcomes. Thus, by recognizing that economists and policymakers face Knightian uncertainty, the KUE approach synthesizes the insights contributed by the major advances in macroeconomic and finance theory – REH and behavioral finance – since the 1970s.