In this project, our interest is to develop numerical scheme for a class of BSDEs with weak terminal condition. This class of BSDEs was introduced by Bouchard, Elie and Reveillac [1], in which the terminal value YT of the portfolio is required to satisfy a weak constraint. From a financial point of view, this approach is referred to as quantile or efficient hedging, and was first discussed by F Ìollmer and Leukert [2, 3]. In particular, they explained how the so-called quantile hedging price for European option can be computed explicitly in a complete market, using duality arguments and Neyman-Pearson lemma.
References
[1] Bruno Bouchard, Romuald Elie, and Antony Rveillac. Bsdes with weak terminal con- dition. The Annals of Probability, 43(2):572â604, 2015.
[2] Hans F Ìollmer and Peter Leukert. Quantile hedging. Finance Stoch., 3(3):251â273, 1999.
[3] Hans F Ìollmer and Peter Leukert. Efficient hedging: cost versus shortfall risk. Finance Stoch., 4(2):117â146, 2000.
The candidate will work on the Le Cam estimation procedure in autoregressive processes driven by stationary Gaussian noise and random coefficient autoregressive processes. The relation between this methodology and recent machine learning procedures will also be discussed during the postdoctorate. Autoregressive processes are relatively common in the analysis of temporal series in insurance.
Particularly, the joint estimation of the drift parameter, variance parameter and Hurst parameter in the autoregressive process driven by the fractional Gaussian noise will be considered. This work follows recent works on the topic:
[1] A. Brouste, C. Cai and M. Kleptsyna (2014) Asymptotic properties of the MLE for the autoregressive process coefficients under stationary Gaussian noises, Mathematical Methods of Statistics, 23(2), 103-115
[2] Marius Soltane (2018) Asymptotic efficiency in the autoregressive process driven by a stationary Gaussian noise, hal-01899971.
[3] A. Brouste, C. Cai, M. Soltane and L. Wang (2020) Testing for the change of the mean-reverting parameter of an autoregressive model with stationary Gaussian noise, Statistical Inference for Stochastic Processes, 23(2), 301-318.
Oleg Chernoyarov is professor of Moscow Power Institute and works in statistical radiophysics. The goal of his visit is to continue the cooperation started 6 years ago in the field of detection and estimation of signals observed in different noises. It is supposed to study stochastic models related with GPS-localization and to describe the errors of estimation in the case of corresponding hidden Markov models .