Education

PhD Thesis (1999) Two Problems in Environmental Statistics: Capture-Recapture Analysis and Smooth Extremal models. Department of Mathematics, EPFL Lausanne.
Undergraduate Thesis (1994) Statistical Analysis of Toxological Data. Department of Mathematics, EPFL Lausanne.
Undergraduate Diploma in Mathematics (1993) Department of Mathematics, EPFL Lausanne.

Employment

Current: Tenure Track Assistant Professor in Statistics, Department of Operations Faculty of Business and Economics University of Lausanne   2009--2011: In charge of the project “Centre interfacultaire de Statistique” at University of Geneva. Lecturer at EPFL teaching “Quantitative Risk Management”, Master in Financial Engineering, Master of Mathematical Sciences. Lecturer at University of Geneva teaching “Multivariate Analysis”, Master of Statistics and “Statistiques et Probabilités I”, Bachelor pour Relations Internationales. 2009--2010: Scientist at EPFL, Institute of Mathematics,  Chair of Statistics. Member of the RiskLab, ETH Zurich. Lecturer at EPFL teaching “Quantitative Risk Management”, Master in Financial Engineering, Master of Mathematical Sciences. Lecturer at University of Geneva teaching “Multivariate Analysis”, Master of Statistics and “Statistiques et Probabilités I”, Bachelor pour Relations Internationales. 2007--2009 Head of Statistics at the Direction of Education, EPFL Lausanne. 2004--2007 Quantitative Analyst/Risk manager, JQS Investment Advisors SA, Switzerland. 2002--2005 Lecturer at the Department of Econometrics, University of Geneva. 2002-2006 NCCR-FINRISK Research Fellow at the Department of Mathematics, ETH Zurich. 2000-2002 Postdoc at the Eidg. Institut für Schnee- und Lawinenforschung, SLF Davos. April 2000 RiskLab Research Fellow at the Department of Mathematics, ETH Zurich. 1996-1999 PhD student in Statistics at the Department of Mathematics, EPFL Lausanne.

Research Interests

Statistical methods for the modelling of extremal events in general, and quantitative risk management in particular. Extreme value theory, dependence modelling, copulas, statistical analysis of point processes, operational risk. Applications in environmental science, insurance and finance.

Teaching

Quantitative Risk Management (2008, 2009,2010) Master of Mathematical Sciences and Master in Financial Engineering, EPFL Lausanne.
Multivariate Data Analysis (2009,2010) Master of Statistics, University of Geneva.
Méthodes Quantitatives (2009) IGUL, University of Lausanne.
Statistiques et Probabilités (2002, 2003, 2004, 2005, 2009,2010) SES, University of Geneva.

 

Scientific Publications

Extreme-quantile tracking for financial time series. To appear in Journal of Econometrics (with P. Embrechts and S. Sardy).

High-frequency financial data modeling using Hawkes processes. 2011. Submitted (with J. McGill).

Discussion of the paper: Threshold modelling of spatially dependent non-stationary extremes with application to hurricane-induced wave heights. Environmetrics, 22(7) (with A.C. Davison and L. Frossard).

An EVT primer for credit risk. 2011. In Lipton, A. and Rennie, A. (Editors) Handbook of Credit Derivatives, 500-532, Oxford University Press (with P. Embrechts).

Operational risk: the advanced measurement approach. 2011. Encyclopedia of Quantitative Finance, John Wiley (with P. Embrechts).

Copulas and dependence concepts in insurance. 2010. Encyclopedia of Quantitative Finance, John Wiley, 379-382 (with P. Embrechts).

Revisiting the edge, ten years on. 2010. Communications in Statistics - Theory and Method, 39, 1674-1688 (with P. Embrechts).

Infinite mean models and the LDA for operational risk. 2006. Journal of Operational Risk. 1(1), 3-25 (with P. Embrechts and J. Neslehova).

Quantitative models for operational risk: extremes, dependence and aggregation. 2006. Journal of Banking and Finance. 30, 2635-2658 (with P. Embrechts and J. Neslehova).

A point process approach to Value-at-Risk estimation. 2005. Quantitative Finance 5(2), 227-234 (with A.C. Davison and A.J. McNeil).

Generalized additive modelling of sample extremes. 2005. Journal of the Royal Statistical Society, C. 54(1), 207-222 (with A.C. Davison).

Smooth extremal models in finance. 2004. Journal of Risk and Insurance. 71(2), 183-199 (with P. Embrechts).

A statistical analysis of the shareprice of the SAIR group (1996-2001) from a risk manager's point of view. 2002. Derivatives Use, Trading & Regulation. 8, 105-122 (with P. Embrechts and A. Roehrl).

Bayesian inference for small-sample capture-recapture data. (1999). Biometrics 55, 727-731.

Working Papers and Other Publications

Nonparametric Copulas modelling for risk measures with nonstationary dependence. 2010. Working paper.

EVT can save your neck. 2004. Bulletin of Swiss Statistical Society (with A.S.A. Roehrl).

Was ist Extremwerttheorie? 2004. Risknews.

Extreme Datamining. 2003. Published in Between Data Science and Applied Data Analysis, Gaul/Vichi/Schader: Springer  (with S.A. Jarvis, R. Perera, A.S.A Roehrl, S.W. Schmiedl, Sondergaard M.P).

Datamining mit R. 2002. Linux-Enterprise (with A.S.A Roehrl, R.A Roehrl, S.W. Schmiedl).

A comparison between l$_1$ Markov random field-based and wavelet-based estimators. 2002. Published in Statistical Data Analysis Based on the L$_1$-Norm and Related Method. edited by Y.Dodge, Boston: Birkhauser (with S. Sardy, C. Bilat, P. Tseng).

Risk reduction: Transparent real-time enterprise. 2002. Banks and Technologies (with A. Weinberg, V. Berezka, A.S.A Roehrl, S. Schmiedl).

Projekt "ZUP": Erstellen einer transparenten und effizienten real-time Firma. 2002. Linux-enterprise (with V. Berezka, S. Jarvis, V.  Kuminov, V. Lenshin,  A. Roehrl, S.  Schmiedl, A. Weinberg).

Durchblick, Was macht Usability so brauchbar? 2002. Linux-Enterprise (with A. Roehrl, S. Schmiedl, R. Kalavasis, A. Maines).

R - Un exemple du succès des modèles libres. 2001. Flash Informatique, Swiss Federal Institute of Technology, Lausanne (with D. Kuonen).

Technique de lissage pour excès au-delà d'un seuil.  1999. Published in XXXIèmes Journées de Statistique, 179.

Etude du paramètre de survie dans un problème de capture-recapture: ajustement de la vraisemblance profile et approche par méthodes bootstrap. 1998. Technical Report,  98.,  Department of Mathematics, Swiss Federal Institute of Technology, Lausanne.

Ajustement de vraisemblance selon McCullagh et Tibshirani. 1997. Technical Report, 97.1, Department of Mathematics, Swiss Federal Instituteof Technology, Lausanne.

Test de permutation de l'indépendance dans un tableau de contingence 2X2. 1995. Technical Report, 95.1, Department of Mathematics, Swiss Federal Institute of Technology, Lausanne.

Développement de Méthodes Statistiques pour l'Analyse de Données Toxicologiques. 1994. Undergraduate Thesis, Department of Mathematics, Swiss Federal Institute of Technology, Lausanne.

Thesis Supervision

PhD co-examiner and master thesis supervisor. I have further supervised several theses at the Bachelor level.

PhD Thesis

Chavez-Demoulin, V. (1999) Two Problems in Environmental Statistics: Capture-Recapture Analysis and Smooth Extremal models. Ph.D. Thesis. Department of Mathematics, EPFL Lausanne.

Book Review

Chavez-Demoulin, V. (2009) Finite Mixture and Markov Switching Models. S. Frühwirth-Schnatter. Springer, New York. JASA, March 2009, Vol. 104, No. 485.

Referee Work

Biometrics, Biometrika, Journal of Econometrics, Insurance: Mathematics & Economics, ASTIN BULLETIN, Computational Statistics & Data Analysis and Statistics.

Talks

ORX Analytics Forum, Turin, November 2010.

CFIES’2010 : Deuxième colloque francophone international sur l’enseignement de la statistique, Université libre de Bruxelles, septembre 2010.

Climate Change and Extreme Events: Managing Tail Risks, at the "Resources for the Future" institute in Washington DC, February 2010.

Computational Management Science, University of Geneva, May 2009.

OpRisk Europe 10th Annual Conference, London, April 2008.

Cherry Bud Workshop, Keio University, Tokyo (Japan), March 2007.

Séminaire de Statistique, University of Geneva, April 2006.

Taught a course on Extreme Value Theory and Dependence Modelling (together with P. Embrechts and J. Neslehova) at the Federal Reserve Bank of Boston, September 2005.

Workshop on Risk Analysis and Extreme Values, Department of Statistics of Paris VI, June 2005.

Implementing an AMA for Operational Risk, Federal Reserve Bank of Boston, May 2005.

International Meeting on Extreme Value Analysis, University of Aveiro, July 2004.

Swiss Statistics Meeting, Montreux, October 2003.

Aon Re Europe, Science Team Meeting, EURANDOM, Eindhoven University of Technology, September 2003.

Séminaire du Département d'Econométrie, University of Geneva, April 2003.

Coupling Climate and Economic Dynamics, NCCR-Climate, WBCSD Geneva, November 2002.

Risk Day, ETH Zürich, October 2002.

NCCR FINRISK Site visit, University of Zurich, October 2002.

Séminaire de Probabilité-Statistique, Université de Marne-la-Vallée, May 2002.

ETH Zürich, January 2002.

International Meeting on Extreme Value Analysis, Catholic University of Leuven, June 2001.

Eidg. Institut für Schnee- und Lawinenforschung, SLF Davos, January 2001.

3ème Cycle Romand de Statistique, Villars-sur-Ollon, February 2000.

31èmes Journées de Statistique, Grenoble (France), May 1999.

Sixth Valencia International Meeting on Bayesian Statistics, Alcossebre (Spain), May 1998.

Membership

Swiss Statistical Society.

Statistical Software Development

The Riskometer grew out of the NCCR FINRISK research project in collaboration with the statistical software (S-PLUS) provider Insightful. We developed an interactive tool for the statistical estimation of quantitative risk measures based on time series data. Examples include quantile-based risk measures like Value-at-Risk (Return Periods) and Expected Shortfall (Mean Residual or Excess Life). The models included are based on past and current research in Extreme Value Theory (EVT) for time series models. Through backtesting procedures, the new, EVT-based methodology is compared and contrasted with more classical models strongly based on Gaussian assumptions. Though the data used for this project come from finance, the S-PLUS modules developed have a much wider range of application.

 

A personal statement

Classical Ballet is to Dance what Mathematics is to Science.