Keynote speakers of the JEF'2020 Conference
Professor Jean-Marc Bardet
Université Paris 1 - Panthéon-Sorbonne, France
Short Bio: Jean-Marc Bardet is a full professor of applied mathematics in University Paris 1-Panthéon-Sorbonne since 2003. He completed his Ph.D. at the University Paris Sud-Orsay in 1997 under the supervision of D. Dacunha-Castelle. He also was an assistant professor in University Toulouse 3. His main thematics of research is "limit theorems, statistics and applications for time series and stochastic processes". He has published several articles in Annals of Statistics, Bernoulli, Stochastic Processes and Applications, Electronic Journal of Statistics, Journal of Time Series Analysis, Journal of Multivariate Analysis ...
Title: Contrast estimation of general locally stationary processes using coupling
Summary: This talk aims at providing statistical guarantees for a kernel based estimation of time varying parameters driving the dynamic of very generals classes of local stationary processes. We consider coupling arguments in order to extend the results of Dahlhaus to the local stationary version of the infinite memory processes. The estimators are computed as localized M-estimators of any contrast satisfying appropriate regularity conditions.
We prove the uniform consistency and pointwise asymptotic normality of such kernel based estimators. We apply our results to usual contrasts such as least-square, least absolute value, or quasi-maximum likelihood contrasts. Various local-stationary processes as ARMA, AR(∞), GARCH, ARCH(∞), ARMA-GARCH, LARCH(∞) …. and integer valued processes are also considered. Numerical experiments demonstrate the efficiency of the estimators on both simulated and real data sets such as log-returns of financial index.
Professor Frederi Viens
Michigan State University, USA
Short Bio: Dr. Frederi Viens is a Professor of Statistics and Probability at Michigan State University, USA. He wrote his Ph.D. in probability theory at the University of California at Irvine, graduating in 1996. He has published nearly 90 articles in probability and stochastic processes, mathematical statistics, finance and insurance mathematics, and Bayesian statistics, with applications in diverse areas from climate science to agricultural economics, nuclear physics, agronomy, and to hospital management. He is associate editor of seven journals in probability, statistics, and finance. He is a fellow of the Institute of Mathematical Statistics, is scientific moderator for the Seminar on Stochastic Processes, and has held fellowships from the Fulbright commission and the US Department of State. He is funded by the National Science Foundation, the Office of Naval Research, and the US Department of Agriculture.
Title: Fractional Brownian models in finance, and option pricing with a martingale method for path-dependent PDEs
Summary: The first half of this presentation will provide a brief introduction to a continuous-time framework for high-frequency stochastic volatility modeling in finance, including the use of fractional Brownian motion (fBm), and how it relates to rough volatility and long-memory models. These processes, which are highly non-Markovian, defy standard attempts at computing objects such as hedging-based option prices. In the presentation’s second part, we will provide a brief overview of a new technique for path-dependent PDEs, which allows to express option prices as conditional expectations based on the entire path of a martingale component of the underlying fBm-type process. In a dynamic setting, this requires a concatenation of the observed path up to the current time with a specific observable curve derived from the distribution of the future paths. The application to option pricing and hedging in a financial market with rough volatility, uses critically the fact that forward variances are quoted in the said market. (This is joint work with Jianfeng Zhang from the University of Southern California).