Financial Systemic Risk Measures based on Monte Carlo Simulation: Theory and Methods

Description

Systemic risk has been one of the most important issues in financial risk management, and plays a central role in regulatory frameworks of financial systems. As a useful tool in measurement of systemic risk, Monte Carlo simulation allows from complex structure of systemic models, which is appealing in practical applications. However, application of Monte Carlo simulation to systemic risk measurement is far from straightforward, and research in this area has been underdeveloped. This project aims to fill this gap.In this project, we propose to study Monte Carlo methods for systemic risk measures, including the commonly used conditional value-at-risk (CoVaR) and marginal expected shortfall (MES). We focus on the development of efficient simulation methods and variance-reduction techniques that produce efficient estimators for these systemic risk measures, and their sensitivities. With the proposed methods for sensitivity analysis, we propose to further study the quantification of model uncertainty in systemic risk measurement, portfolio optimization under systemic risk constraints and approaches to distributionally robustifying these models. The project focuses on both the design of practically useful methods and algorithms, and their theoretical analysis. It is expected that research outputs of this project may lead to a set of useful quantitative tools for systemic risk measurement with sound theoretical guarantees.