Martin Ellison, University of Oxford: "Techniques and applications of robust control"

Course outline:

The aim of the course is to introduce participants to the central techniques and methods used in applying robust control to economic problems. The motivation for doing so is to provide a disciplined response to economic agents fearing that the framework they are using to make decisions may be misspecified. The theory has obvious attraction for macroeconomic policymakers concerned that they may not perfectly understand the workings of the economy, but can equally well be applied to consumers, firms or unions making decisions in an environment of unstructured uncertainty. By the end of the course participants will be able derive robustly optimal policies in a wide variety of environments.

The course is split over two days. The first day is spent on the philosophy and theoretical underpinnings of robust control, starting from basic methods and progressing through a discussion of entropy and multiplier constraints to worked examples in a set of static and dynamic models. The first day concludes by introducing detection error probabilities as a way of calibrating an agent’s fears of misspecification. The second day presents a series of papers that apply robust control techniques to real world economic situations. In the morning the focus is on macroeconomic policymaking, whereas the afternoon looks at applications of robust control to finance.

Target audience:

The course is targeted at students at any stage of their PhD studies. The instructor will assume knowledge of standard postgraduate macroeconomics at the level of Romer (2011) Advanced Macroeconomics, but otherwise all techniques will be explained from first principles. The course is not specifically aimed at macroeconomists, so should also be useful for students interested in finance or microeconomics (where fear of misspecification is typically referred to as ambiguity aversion). A familiarity with simple techniques in dynamic programming is helpful but not essential.

Length of course:

2 days, 5 hours per day of lectures

Background reading:

Barillas, B., Hansen, L.P. and Sargent (2009), ‘Doubts or variability?’, Journal of Economic Theory 144, 2388‐2418

Ellison, M. and T.J. Sargent (2012), ‘A defence of the FOMC’, International Economic Review, forthcoming

Ellison, M. and T.J. Sargent (2012), ‘The welfare costs of business cycles in robust economies with individual consumption risk”, mimeo

Hansen. L.P. and T.J. Sargent (2008), Robustness, Princeton University Press