chair of entrepreneurial risks professor: didier sornette

Didier Sornette joined ETH Zurich in March 2006. Professor Sornette (h-index 49 (ISI) and 67 (scHolar)) graduated from Ecole Normal Supérieure Ulm, , in 1981 and became a Research Director at CNRS (1981-2006); from 1996, he was jointly Professor at the Institute of Geophysics and Planetary and the Department of Earth and Space Sciences at UCLA (1996-2006). While holding the Chair of Entrepreneurial Risks at ETH Zurich, he is also a Professor of Finance at the Swiss Finance Institute, a Professor of Physics associated with the Department of Physics (D-PHYS) at ETH Zurich, and a Professor of Geophysics associated with the of fi nancial bubbles and crashes, and the diagnosis of sys- Department of Earth Sciences (D-ERDW) at ETH Zurich. In temic instabilities. Other areas of work include earthquake addition, he is an Honorary Professor at the School of Busi- physics and geophysics, fi nancial and the theory ness and Research Center of Systems Engineering, East Chi- of complex systems, the dynamics of success on social net- na University of Science and Technology (ECUST), Shanghai, works, and the complex systems approach to medicine (im- China. He is a founding member of the Risk Center, the com- mune system, human parturition, epilepsy, etc.). In 2008, he petence center that plays a central role in ETH Zurich’s Glob- launched the Observatory to test the hy- al Risk Initiative. His research focuses on the prediction of pothesis that fi nancial bubbles can be diagnosed in real crises and extreme events in complex systems, in particular time and their termination predicted probabilistically.

mission statement of Group

Our group belongs to the “Systems Design and Risk” re- resilient societies (recent advances in this direction in- search pillar with D-MTEC and is working on a variety of clude the Financial Crisis Observatory and the endo-exo projects related to entrepreneurial risks. Specifi cally, we theory of shocks); view the world as an ensemble of entangled and inherently (ii) develop a general theory of human preferences and de- out-of-equilibrium networks. We strive to push the limits of cision making at the individual level and taking account silo disciplines by developing a combined inter-, multi- and of social interactions (recent progress includes the for- transdisciplinary view of knowledge, using both fundamen- mulation of a new “Quantum Decision Theory” that tal research and operational implementation, as well as en- solves – consistently and without adjustable parame- trepreneurial initiatives. The multidisciplinary approach is ters – all the paradoxes and fallacies known to plague refl ected in the multiple academic associations D. Sornette available decision theories, including Kahneman and has with different departments. Twersky’s Prospect Theory); Our main objectives are to (iii) transfer our research and knowledge to industry and so- (i) develop, validate, and transfer into best practice a gen- ciety via collaborations, joint ventures, and operational eral theory for the understanding, characterization, implementation. prediction, and eventual control of crises to build more

73 www.er.ethz.ch

Group and Resources

In addition to Didier Sornette (DS), our group is constituted - Yaver Kamer is developing risk models for earthquakes as follows: and testing their predictability; (i) Four senior researchers: Dr. Peter Cauwels and Dr. Ryan - Tatyana Kovalenko is studying cognitive limits in decision Woodard are the core members of the Financial Crisis making to test and extend the quantum decision theory Observatory (FCO); Prof. Alexander Saichev develops, with developed by Didier Sornette and Prof. V. Yukalov; Didier Sornette, theoretical models of volatility, self-excited - Susanne von der Becke is working on a general framework cascades and risks; Dr. Monika Gisler co-heads the project for understanding and controlling systemic risks for finan- on “Social Bubbles” with Didier Sornette; cial systems, with particular emphasis on the role of the (ii) Two postdocs: Dr. Vladimir Filimonov co-invented with banking sector; Didier Sornette the self-excited multifractal process for fi- - Yaming Wang is developing methods for fault network re- nancial volatility and is co-discoverer of a new method for construction, based on past seismicity including location precisely quantifying in real time the endogeneity level of uncertainties, in order to better localize the future occur- financial markets, with broad applications to many systems. rence of earthquakes; Dr. Thomas Maillart, who did his PhD under Didier Sornette, - Qunzhi Zhang is quantifying the impact of news on finan- is now a part-time postdoc with our group working on a cial market moves and is the main architect of the reverse- web-based implementation of collective action called engineering project using agent-based models. Risikopedia; Our group benefits from collaboration with regular visiting (iii) 11 PhD students: professors (Taisei Kaizoji, Y. Malevergne, V. Yukalov, and - Maroussia Favre is developing models of human coopera- E. Yukalova), as well as exchange students, postdocs and tion and gender difference in risk taking; professors from China and Russia, in particular. - Zalan Forro is working on the valuation of social compa- The group has developed a strong computational infra- nies (YouTube, Groupon Zynga) in the Financial Crisis Ob- structure, which serves the operations of our Financial Crisis servatory and on GIS-bubble approaches to the real estate Observatory, among others. Our core infrastructure consists market in Switzerland, using Comparis AG’s unique data- of three types of hardware: personal computers and lap- base; tops (for quick, interactive analysis), a private rack-mounted - Georges Harras is finishing his thesis on models of herd- cluster of five nodes, each containing four quad-core pro- ing behavior and abnormal correlations in financial mar- cessors (a total of 80 cores for medium-sized analysis and a kets; considerable amount of post-processing) and the ETH Zu- - Ryohei Hisano is working on a large database, involving in rich Brutus cluster. Our ETHZ-FCO is the largest single share- particular millions of companies in Japan, to test models holder of Brutus, having purchased over 12% of the process- of growth and competition; ing power of this shared-resource cluster of 10,000 cores. - Andreas Huesler is working on the agency problem in the Brutus is used for the main daily scans and analyses of over mutual fund industry, on endogenous two thousand global assets. The private cluster and Brutus models and their calibration for CO2 production, and on have been designed with large memory and disk storage, the data analysis of financial bubbles in the laboratory; high speed internal networking and daily RAID data backup. - Mohamed Issami is working on coding trading strategies for agent-based models and the problem of reverse-engi- neering in financial markets;

74 chair of entrepreneurial risks

Personnel Data FTE (Ø 2006 - 2011)* Financial Data (Expenditures) CHF (Ø 2006 - 2011)*

Professor 1.0 ETH Basic 1‘210‘000 Funding Scientifi c Staff with 3.4 Competitive / Additional 375‘000 Doctoral Degree External SNSF 62‘000 Scientifi c Staff without 5.5 Funding Doctoral Degree CTI 67‘000 (mainly PhD Students) EU & Intl. Org.

Student Assistants 0.8 Federal Offi ces 2‘000

Admin. / Techn. / IT Staff 1.5 Private Sector 124‘000

Total 12.2 Total 1‘840‘000

* Average of full years at ETHZ. tablE 1: PErsonnEl and finanCial data of thE Chair of Prof. sornEttE.

75 www.er.ethz.ch

Research Activities and Achievements

Financial Bubbles and Crashes: Our first priority is to use which are activated intermittently. This gives rise to specific the Financial Crisis Observatory (http://www.er.ethz.ch/ properties and signatures that may be the unique charac- fco/) to test the hypotheses that (i) financial bubbles can teristics of dragon-kings, thus leading to possible predic- be diagnosed in advance before they burst and (ii) their ter- tability. We have edited a special volume on dragon-kings mination time can be predicted probabilistically much bet- soon to appear with Springer, compiling 20 contributions in ter than chance. In 2008, we launched the financial bubble different scientific fields. Overall, the picture that emerges experiment (http://fcofbe.blogspot.com/) and the financial is that dragon-kings reveal hidden mechanisms, which are trading experiments to test the existence of predictability only transiently active and which amplify normal fluctua- and inefficiencies in financial markets empirically. We have tions (often described by the power laws of the normal re- also developed a series of mathematical models, agent- gime). The dragon-king worldview also provides a general based simulations, and a large toolbox of methods to repre- approach for understanding the present world financial sent and characterize the dynamics of financial prices and, crisis as based on two decades of successive financial and in particular, their most extreme expressions in the form of economic bubbles, inflating the Mother of All Bubbles, with bubbles and crashes. new monster dragon-kings on the horizon. The Endo-Exo Dynamics of Success, Fame, and Crises: We Our research contributions are broken down into the found that in many systems, large dynamic changes result following topics (see http://www.er.ethz.ch/publications/ from the entanglement of endogenous and exogenous index): shocks. We have developed a general theory describing the generic dynamic law that characterizes the activity of sys- 1. Social Systems/Finance tems leading to and following a peak, a crisis or a climax. 1.1 Bubbles and Crashes: Theory A profound relationship was found between the response 1.2 Bubbles and Crashes: Theory - Empirical Analyses to exogenous shocks and the endogenous fluctuations of 1.3 Social Bubbles systems driven by epidemic-like interactions, such as word- 1.4 Methods and Techniques for Financial Markets of-mouth contacts in social networks and the triggering 1.5 Agent-Based Models processes of natural systems. This constitutes a generalized 1.6 Asset Pricing, Portfolio Theory, Option Theory, Yield Curve fluctuation-susceptibility theorem for such out-of-equilib- 1.7 Decision Theory, Behavioral Finance, Social Dilemmas, rium systems with punctuated dynamics. Applications in- Cooperation clude the dynamics of commercial sales, YouTube video suc- 1.8 Volatility and Return, Anomalies and Predictions cesses, financial volatility shocks, market crashes, outbursts 2. Complex Systems and Self-Organization of cyber-risks, social conflicts and crises, epileptic seizures, 2.1 Predictability and Endo-Exo Dynamics earthquakes, landslides, climate dynamics, and so on. 2.2 Internet, Open-Source Software, Cyber-Risks, Networks Dragon-Kings: Extreme fluctuations or events are often 2.3 Epileptic Seizures, the Immune System, Biological Sys- associated with statistics. Indeed, power law tems tails are often seen as the epitome of extreme events (the 2.4 Concepts and Techniques “Black Swan” story). We have introduced a different view of 2.5 Self-Organized Criticality, Critical Systems, Power Laws the world of risks, captured by the term “dragon-king.” We 2.6 Self-Excited (Hawkes) Branching Processes claim that extreme events in most natural and social sys- 3. Discrete Scale Invariance & Complex Dimensions tems may not be due to the same mechanisms that act on 4. Earthquakes, Landslides, Glacier Collapses and Other other events and may require specific amplifying processes, Natural Hazards

[1] crane, R. and D. Sornette, 2008. Robust dynamic classes revealed by meas- [6] Jiang, Z.-Q., W.-X. Zhou, D. Sornette, R. Woodard, K. Bastiaensen and P. Cau- uring the response function of a social system, Proc. Nat. Acad. Sci. USA 105 wels, 2010. Bubble Diagnosis and Prediction of the 2005-2007 and 2008- (41), 15649-15653. 2009 Chinese stock market bubbles, J. Econ. Behav. Organization 74, 149-162. [2] darcet D. and D. Sornette, 2008. Quantitative determination of the level of [7] yukalov, V.I. and D. Sornette, 2011. Decision Theory with Prospect Interfer- cooperation in the presence of punishment in three public good experi- ence and Entanglement, Theory and Decision 70, 283-328. ments, Journal of Economic Interaction and Coordination 3, 137-163. [8] cederman, L.-E., C. Warren and D. Sornette, 2011. Testing Clausewitz: Na- [3] maillart, T., D. Sornette, S. Spaeth and G. von Krogh, 2008. Empirical Tests of tionalism, Mass Mobilization, and the Severity of War, International Or- Zipf’s law Mechanism In Open Source Linux Distribution, Physical Review ganization 65 (4), 605-638. Letters 101, 218701. [9] harras, G. and D. Sornette, How to grow a bubble: A model of myopic [4] yukalov, V.I., D. Sornette, E.P. Yukalova, 2009. Nonlinear Dynamical Model of adapting agents, J. Economic Behavior and Organization 80 (1), 137-152. Regime Switching Between Conventions and Business Cycles, J. Economic [10] Gisler, M., D. Sornette and R. Woodard, 2011. Innovation as a Social Bubble: Behavior and Organization 70, 206-230. The Example of the Human Genome Project, Research Policy 40, 1412-1425. [5] Schweitzer, F., G. Fagiolo, D. Sornette, F. Vega-Redondo, A. Vespignani, and D. R. White, 2009. Economic Networks: The New challenges, Science 325, 422-424.

76 Chair of Entrepreneurial Risks

Year ISI Peer-Reviewed Publications Other Publications by Group Talks Number of Publications Number of Publications ( 2006 - 2011 ) of Group Number of Publications (whole Career) Professor by Group‘s Cited, Times ( 2006 - 2011 ) Publications Professor‘s Cited, Times (whole Career) Publications Book Chapters, Books, Monographs Non-ISI Peer-Reviewed Publications Miscellaneous Members Group by Talks All

2006 27 372* 7 6‘300 3 1 16

2007 14 14 33 33 0 0 24

2008 15 15 64 64 0 0 26

2009 20 20 96 96 1 1 35

2010 16 16 128 128 1 1 36

2011 23 23 177 177 0 3 33

Total 115 460 505 6‘798 5 6 170

* These fields contain the data from the start of the professor‘s career until (and including) 2006. table 2: publications and talks of the chair of prof. Sornette.

Examples of large-scale financial bubbles that have been diagnosed ex-ante and predicted by Sornette‘s group in advanced publications.

77 www.er.ethz.ch

Teaching Activities

Accounting for Managers (Master level taught by Prof. J.-P. vanilla and exotic options with typical models and their Chardonnens) ~ 200 students: Financial accounting: bal- limitations described in several case studies. ance sheet, income statement, recording business transac- Corporate Finance (Master level taught by Dr. Markus R. tions, evaluation. Management accounting: full and partial Neuhaus, CEO PwC Switzerland) ~30 students: Investment costing systems, management decisions. management, value based management, financial report- Financial Management (Master level taught by Prof. J.-P. ing, mergers and acquisitions, internal control, turnaround. Chardonnens) ~ 200 students: Financial objectives, finan- Discovering Management (Bachelor and Master level taught cial statement analysis, working capital management, capi- by 11 professors, Didier Sornette contributing a lecture on tal budgeting techniques, capital structure, cost of capital. Risk Management) ~40 students: Discovering Management Financial Market Risks (Master level taught by Prof. D. Sor- (formerly known as Discovering Entrepreneurship); a gen- nette) ~ 120 students: Financial risks and management, eral introduction to the field of entrepreneurship and the financial markets, options and derivatives, valuation of practices of professional business management. In his lec- bonds, management of international risks. ture, Prof. D. Sornette introduces the students to fat tail Entrepreneurial Risks (Master and PhD levels taught by Prof. risks, cascade effects and systemic risks through case stud- D. Sornette) ~ 50 students: Risks and entrepreneurship, ies and draws conclusions for best risk management prac- power law risks, measure of risks, mathematical models, tice. (http://www.entrepreneurship.ethz.ch/education/lec- origins of crises, market crashes, predictability, human- tures/dm/) made risks. Interdisciplinary Risk Center Weekly Seminar Series: Com- Equity Derivative Trading in Practice (Master and PhD levels plex Socio-Economic Systems and Integrative Risk Manage- taught by Dr. Mika Kastenholz, Head of Credit Suisse Deriva- ment (PhD level co-organized and taught by the 10 mem- tives Trading Desk) ~ 50 students: Filling in the gaps be- bers of the ETH Zurich Risk Center): This course consists of tween textbook quantitative finance and industry practice, presentations and subsequent discussion on modelling emphasis on how theory and quantitative models are used complex socio-economic systems and crises. in derivatives trading on a daily basis, single and multifactor

D-MTEC students in the headquarters of Zurich Financial Service with World Group CEO, James J. Schiro in the center of the picture, during a special week session in 2008 on „Entrepreneurial Leadership“ organized by Prof. Sornette and his assistants.

78 Chair of Entrepreneurial Risks

Services to the Academic Community

The group co-organized the CCSS International Workshop (2006-present). He is an Associate Editor for the Journal of on Coping with Crises in Complex Socio-Economic Systems, Quantitative Finance (2001-present), the International Jour- ETH Zurich, June 8-12, 2009, https://www.soms.ethz.ch/ nal of Modern Physics C (computational physics) (2005-pre- workshop2009/, and June 20-24, 2011 (http://www.ifb.ethz. sent), the Journal of Economic Interaction and Coordination ch/comphys/conferences/ccss-workshop2011/index). Didier (JEIC) (2006-present), Chaos, Solitons and Fractals (2010-pre- Sornette was a distinguished fellow of the Institute of Ad- sent), the European Journal of Finance (2010-present), the vanced Study, Durham University, UK (October 2007-Febru- International Journal of Portfolio Analysis & Management ary 2008) http://www.dur.ac.uk/ias/ and is Honorary Pro- (IJPAM) (2011-present) and the International Journal of Ter- fessor of the East China University of Science and raspace Science and Engineering (IJTSE), published by World Technology, Shanghai, China (since 2009). Didier Sornette is Scientific (April 2009-present). He is a member of the a member of the editorial board of the Society of Economic Springer Lecture Notes in Physics Editorial Board (2001-pre- Science with Heterogeneous Agents (ESHIA) (http://www. sent), Springer Series on Synergetics Editorial Board es-hia.org) and of its official journal, theJournal of Economic (2001-present), Springer Complexity Programme Editorial Interaction and Coordination (JEIC), member of the Global Board (Jan. 2007-present) and of the Scientific Committee Advisory Board of Human Dignity and Humiliation Studies on Risques-Les Cahiers de l’Assurance, 9 rue d’Enghien, (http://www.humiliationstudies.org/), and member of the 75010 Paris, France. Didier Sornette serves on the ETH advisory board of the Interdisciplinary Center for Complex Research Commission (since 2007). Systems (IZKS) of the University of Bonn, Germany

Knowledge and Technology Transfer

The group receives a lot of media attention. Didier Sornette As a founding member of the ETH Competence Center for and the senior members of the group (Monika Gisler, Peter “Coping with Crises in Complex Socio-Economic Systems” in Cauwels and Ryan Woodard) receive weekly (and often dai- 2008 (http://www.ccss.ethz.ch/), which became the Risk ly) requests for interviews from some of the major journals Center in 2011 (http://www.riskcenter.ethz.ch), the group in Switzerland (Neue Zürcher Zeitung (NZZ), Handelszei- contributes to scientific activities, science leadership, and tung, Das Magazin, L’Agefi, NZZ am Sonntag), Germany (Der outreach to the private sector. In particular, we give talks to Spiegel, Finanzwelt, Handelsblatt, EURO am Sonntag), different industries and outlets to promote the Risk Center’s France (Les Echos), the UK (Physics World, New Scientist) scientific vision. and the US (Wall Street Journal, Financial Times, The New York Times, Bloomberg Business week, Technology Review We have on-going collaborations with numerous institu- MIT, ScienceNews) as well as from radio and TV stations. tions and universities worldwide, in particular in the USA, For a selected list of main media outlets, see http://www. Russia, China, France, and Japan. er.ethz.ch/interviews/index.

We are involved in numerous events and conferences to transfer knowledge to the finance and insurance industry. A selected list can be found at http://www.er.ethz.ch/pres- entations/index.

79 www.er.ethz.ch

Outlook

In line with the above “Mission Statement,” the group has financial crises are knowable, can be diagnosed in advance, many projects in development. Perhaps the most ambi- and can be timed probabilistically. We are eager to work tious is our contribution as leader of the new Observatory with central banks and regulators, in collaboration with our and Exploratory of Economic and Financial Crises within colleagues within D-MTEC, in particular, and at ETH Zurich’s the equally ambitious FuturICT project, which is currently a Risk Center, to change the present state of policymaking forerunner in the European Union’s Flagship competition and regulation into a well-functioning financial system to achieve 1 billion Euros of funding over a decade. Indeed, that serves the economy and increases social welfare. the Observatory and Exploratory, which will be coordinat- We also plan to step up our research efforts on decision ing and federating the research of dozens of European re- making and collective action, integrating the many disci- search groups and institutions, has been inspired by our plines involved into a coherent theoretical framework with own FCO. In addition, this model, started in 2008 at ETH operational goals, to help break the stalemate in the pres- Zurich in our group, is now the template for other Observa- ently inadequate description of how boundedly rational hu- tories and Exploratories planned within FuturICT, from bio- mans allocate their resources and steer their lives, compa- logical risks to crimes. nies, and societies. We believe that the challenge of the next We are also excited by the prospect of expanding our Fi- two decades will create extraordinary tests of our ability to nancial Crisis Observatory to validate the hypothesis that adapt to achieve genuine sustainable development.

80 D-MTECPeer Review 2012 Self - Evaluation

Professorships of the Department of Management, Technology, and Economics

Departement Management, Technologie und Ökonomie Departement Management, Technologie und Ökonomie Department of Management, Technology, and Economics www.mtec.ethz.ch Department of Management, Technology, and Economics