Econometric Approaches for Risk Modelling – Econom&Risks

The Econom&Risk project brings experienced researchers in the field of financial econometrics, issued from three universities. It has three main objectives: (1) to propose new statistical methods for estimating risk measures based on conditional moments, particularly the Value-at-Risk (VaR), (2) to propose new procedures validation of VaR and (3) to contribute to the econometrics of liquidity risk. In each of these areas, the members of the Econom&Risk project have proposed different solutions that have been the subject of several leading international publications.
The Econom&Risk project aims at promoting a reproducible research in financial econometrics. To do this, we have proposed two internet platforms, named Exec&Share and RunMyCode. Both websites are based on the innovative concept of «companion website« of scientific publication. A companion website is a website from which users can download and run online the codes and data associated with a scientific publication.

This project has enabled the development of an interactive website that allows to reproduce online the results of scientific articles. This site intended for professionals and the academic community enables users to benefit from the results of our research in a simple way. Designed as a SaaS (Software as a Service), this website allows to replicate online our methods (tests, estimation methods etc.).
The Econom & Risk project resulted in several publications in prestigious international academic journals in econometrics (Econometrica, Econometric Theory, Journal of Econometrics, Journal of Financial Econometrics, etc.) or financial (Review of Finance, Journal of Banking and Finance , Journal of Empirical Finance, etc.). This project has also led to the organization of an international conference and 11 special sessions at international conferences.

The Econom&Risk project, with the financial support of the ANR, allowed us to achieve many results in terms of international publications, of organization and participation in conferences, and the creation of innovative websites for promoting a reproducible research in our discipline.

The Econom&Risk project enabled the organization of an international conference (Orleans, October 20-21, 2011) brings together more than 120 participants from over 15 countries, and 11 special sessions at international conferences. The Econom&Risk project also allowed researchers and doctoral students to participate in numerous national and international conferences (more than 40) including the ESEM (2011, 2012 and 2013), Computational and Financial Econometrics (2011, 2012 and 2013), etc. Finally, the project has resulted in the creation of two websites. RunMyCode Exec & Share websites host a hundred of companion websites and receive nearly 7,000 visits per month. The project RunMyCode was presented to the Google Faculty Summit (September 2012, London) and in several international conferences (First Open Economics Workshop, Cambridge in December 2012, R^3 Meet the Reproducible Research, Orleans, 2012, Analyzing and Improving Collaborative eScience with Social Networks Workshop IEEE e-Science 2012, Chicago). The project was presented in the fall of 2013 at the Salon de la Valorisation du CNRS.

The aim of the project Econom&Risk (Econometric Approaches for Risk Modelling) is to promote a better analysis of financial risk, not only under normal market conditions, but also in the context of financial crisis. The project is focused on two dimensions of financial risk: (i) market risk and (ii) liquidity risk.
The measure of financial risk is one of the main research topics of financial econometrics. This extremely active domain of research has been initiated by the necessity for the banks to compute a risk measure, called the Value-at-Risk (VAR). Given their simplicity and generality, the VaR and the other risk measures for extreme risks, such as the expected shortfall, have revolutionized the management of financial risks and become a common language to compare the risk of different markets, different countries, different portfolios etc. Recent advances in time series analysis have led to the construction of new models, new estimation procedures and new validation methods for modelling volatility of financial markets. In this project, extensions of the previously mentioned developments are proposed in view of applications to the estimation and validation of risk measures.
Besides, we will also consider the liquidity risk. Up until recently, the liquidity of financial assets has typically been viewed as a second-order consideration in the asset-management industry. Liquidity was frequently associated with simple transaction costs that impose little effect, temporary if any, on asset prices and whose shocks could be easily diversified away. Yet, the evidence, especially the recent liquidity crisis that started in August 2007, suggests that liquidity is now a primary concern. In this context, one of the goals of our research effort is to develop methods of identifying and predicting liquidity events, as well as to design trading strategies to hedge these types of risk.
The project Econom&Risks gathers researchers with a strong background in the field of financial econometrics from three Universities and has four main objectives:
- to construct new statistical methodologies for estimating risk measures based on conditional moments;
- to propose semi-parametric methods for VaR estimation;
- to propose new backtesting procedures able to evaluate the accuracy and the validity of risk measures, especially for VaR forecasts.
- to contribute to the econometrics of liquidity risk.
An important element of the project will be the development of an interactive website devoted to the econometrics of financial risks that will allow the professional and the academic community to beneficiate from the results of our research. This website will be developed as a SaaS (Software as a Service) and will enable professional to replicate our methodologies (test-estimation techniques) on their own data in a very simple way, through an Internet service.