Frank Diebold
"It is tremendously important for students and all researchers to continuously think about the difference between what is important and what is merely unsolved. The sweet spot is to address unsolved problems that are simultaneously truly important. To make a powerful contribution, you need to anchor in the market and bring tools to bear on important issues. The programme’s students have the right combination of professional and academic background to do exactly that."
What are you main research areas?
I am active in financial econometrics and forecasting, empirical finance and empirical macroeconomics (especially business cycles). Mostly, I work in applied time-series econometrics, with focus on financial and macroeconomic applications. Almost all my modelling is predictive in one way or another. Yield curve modelling, which I had a chance to teach in the programme, provides a good example of the importance of predictive modelling.
Precisely, what did the course cover?
The course had two related objectives. First to examine yield curve modelling, broadly defined, looking at empirical aspects and the different approaches used by academics and professionals, so as to give a broad treatment of modern developments in the field. Second to look at the links bridging the yield curve (including volatilities) and macroeconomic fundamentals. Throughout we emphasised asset allocation, risk measurement/management, and asset pricing.
In the first part of the course we introduced traditional finance approaches, which centre on theoreticallyappealing arbitrage-free models but forecast poorly, traditional macroeconomic approaches that admit arbitrage but generate better predictions, and new approaches that allow to combine the best of both worlds. In the second part of the course, we looked at real-time monitoring of business conditions, examined volatility and correlation modelling and forecasting. The background to this exploration was a presentation of yield curves as state-space systems and a review of quantitative tools for estimation and prediction, with emphasis on Markov processes.
Did you have apprehensions about teaching in a programme that is opened to professionals and what was your experience like?
I had no reservations about teaching in the programme, in fact quite the opposite. It takes a rare and interesting breed of person to combine work in a highly-demanding industry position with the rigours of doctoral studies. With their level of expertise, engagement, and appreciation for the industryrelevance of what we were doing, the executive track participants brought a huge benefit to the group. The dynamics in that classroom were really quite special; there were lively discussions with participants’ contributing comments and insights grounded on a combination of solid academic background and rich real-world experiences from diverse sectors. Students understood the academic subject matter, were familiar with the markets, and were very excited to use the material.
As part of the programme’s doctoral workshop series, you presented a working paper (Better to Give than to Receive: Predictive Directional Measurement of Volatility Spillovers, co-authored with Kamil Yilmaz) that looks at volatility spillovers; could you tell us more about it?
The big picture is thinking about connectedness; correlation risk is something we may want to control at the individual investor level as well as the systemic level. During crises, financial market volatility typically increases sharply and spills over across markets. Being able to measure and monitor such spillovers could help provide early warning mechanisms and help track the progress of extant crises.
The paper uses a generalised vector autoregressive framework to decompose the uncertainty of an asset’s return into asset-specific components and external shocks, and proposes measures of both total and directional volatility spillovers, the latter allowing us to shed new light on the nature of cross-market volatility transmission.
We use our methods to characterise daily volatility spillovers across U.S. stock, bond, foreign exchange and commodities markets, over a ten-year period ending January 2010. We find that despite significant volatility fluctuations in all four markets over the period, cross-market volatility spillovers were quite limited until the global financial crisis that began in 2007. We observe that as the crisis intensified so too did the volatility spillovers and pinpoint particularly important spillovers from the stock market to other markets taking place after the collapse of Lehman Brothers in September 2008.
You recently co-edited a book1 on the known, the unknown, and the unknowable in financial risk management. Why devote time to such a project?
Because it’s tremendously important! I am one of the Directors of the Wharton Financial Institutions Center, and we are very interested in financial risk management in general. The global financial crisis has underlined a number of issues with risk management and we felt it was the right time to think more deeply about different aspects of the discipline, from the most quantitative, through to the much more nebulous aspects.
What we noticed is that it is important to take into account the whole spectrum or risks–from known, to unknown, and even unknowable–to conceptualise the different kinds of financial risks in a more realistic and holistic framework and design effective strategies for managing them. The risk management literature has focused on known risks and largely ignored the equally relevant unknown and unknowable situations. Perhaps depressingly for those who have focused on knowable risk, many of the “killer risks” that can really bring firms down belong to the realm of the unknown and unknowable. Against this backdrop, the book reveals the strengths and limitations of «quantitative» risk management, but more importantly provides a framework to construct portfolios, contracts, firms, and policies that can better withstand shocks. The book demonstrates that killer risks are often crucially linked to misaligned incentives and suggests mechanisms to solve the principal-agent problems and offer incentives to do the right thing in situations that were hard to imagine ex-ante. I had a great time working on that book.
Why join the EDHEC-Risk Institute PhD in Finance as Affiliate Faculty?
René Garcia, for whom I have immense respect, told me about the programme and asked me to contribute, which I was happy to do. Looking at the other faculty members, it is clear that it is a first-rate programme. Teaching this course has been a wonderful experience; I was stunned by the quality of the students, the premises, and the organisation, and am looking forward to doing this again, perhaps on the Singapore campus.
Would you have recommendations for doctoral students?
It is tremendously important for students and all researchers to continuously think about the difference between what is important and what is merely unsolved. The sweet spot is to address unsolved problems that are simultaneously truly important. To make a powerful contribution, you need to anchor in the market and bring tools to bear on important issues. The programme’s students have the right combination of professional and academic background to do exactly that.
1The Known, the Unknown, and the Unknowable in Financial Risk Management: Measurement and Theory Advancing Practice, edited by Francis X. Diebold, Neil A. Doherty, and Richard J. Herring, Princeton University Press, 2010.
