Ophélie Couperier

Elicitability of Market-Based Systemic-Risk Measures (2022) with Sylvain Benoit, Jérémy Leymarie and Olivier Scaillet
 

Abstract: A risk measure, or more generally a statistical functional, is called elicitable if it can be defined as the minimizer of a suitable expected scoring function, see Gneiting (2011) and references therein. This article explores the notion of elicitability (and identifiability) for systemic-risk measures that are used to identify the financial institutions contributing the most to the overall risk in the financial system. Our elicitation framework applies to systemic-risk measures that are expressed as a function of the expected equity loss conditional on a financial crisis, such as the marginal expected shortfall (MES), the systemic expected shortfall (SES), or the systemic-risk measure SRISK. This property paves the way to the implementation of M-estimation for the systemic-risk measures or to the comparison and backtesting of the systemic risk models used by academics and policy makers to rank the systemically important financial institutions (SIFIs) whose failure might trigger a crisis in the entire financial system.

​

Research grant sponsored by the Fondation Banque de France (30k€), 2021-2022

​

​

Daily volatility forecasting using intraday returns and functional covariates (2022) with Christian Francq and Jean-Michel Zakoïan
 

Abstract: In addition to past daily returns, intraday returns are now well-known to carry valuable information on the daily volatility dynamics. GARCH models with exogenous scalar covariates (like realized volatilities) partially take into account this additional information. We propose a volatility model including functional covariates, for handling the whole information conveyed by the low and highfrequency returns. We start by giving general stochastic results (strict stationarity, existence of moments and log-moments). Then we study estimation of the model by Quasi-Maximum Likelihood (QML) and we propose a portmanteau test of goodness-of-fit. Monte Carlo simulations and an empirical application on financial series illustrate the interest of including functional covariates for volatility prediction.

​

​

Backtesting Expected Shortfall via Multi-Quantile Regression (2021) with Jérémy Leymarie                                                               Code & data

​

Abstract: This article proposes an original approach to backtest Expected Shortfall (ES) exploiting the definition of ES as a function of Value-at-Risk (VaR). Our methodology examines jointly the validity of the VaR forecasts along the tail distribution of the risk model and encompasses the Basel Committee recommendation of verifying quantiles at risk levels 97.5% and 99%. We introduce four easy-to-use backtests in which we regress the ex-post losses on the VaR forecasts in a multi-quantile regression model, and test the resulting parameter estimates. Monte-Carlo simulations show that our tests are powerful to detect various model misspecifications. We apply our backtests on S&P500 returns over the period 2007-2012. Our tests identify misleading ES forecasts in this period of financial turmoil. Empirical results also show that the detection abilities are higher when the evaluation procedure involves more than two quantiles, which should accordingly be taken into account in the current regulatory guidelines.

​

Best paper award at the Annual Meeting of the German Finance Association (DGF), 2019