This talk has been automatically captioned. You can remove these by pressing CC on the video toolbar.

Name: Jingyu Ji

Talk Title: Autoregressive Conditional Accelerated Frechet Model for Decoupling Systemic Risk into Endogenous and Exogenous Competing Risks

Abstract: Identifying systemic risk patterns in social, political, economic, financial, market, regional, global, environmental, transportation, epidemiological, material, chemical, and physical systems and their impacts is the key to risk management. This work integrates the newly introduced new extreme value theory for maxima of maxima and a new time series benchmark model of autoregressive conditional Frechet (AcF) for modeling systemic risk into a new autoregressive conditional accelerated Frechet (AcAF) model for decoupling systemic risk into endogenous and exogenous competing risks. In the paper, the focus is paid on market risk and systemic financial risk. Nevertheless, the AcAF model can be applied to all systems above and beyond. The AcAF model and its resulting competing risks provide clear endogenous and exogenous competing risk patterns of market risks and reveal the causes of the financial crisis, which was not detected using the existing models. The probabilistic properties of stationarity and ergodicity of the AcAF model are established. Statistical inference is developed through conditional maximum likelihood estimation. The consistency and asymptotic normality of the estimators are derived. Simulated numerical examples are used to demonstrate the efficiency of the proposed estimators. Empirical studies of time series of maxima of maxima of stock returns in S&P 500, intra-day returns of high-frequency trading stocks, and intra-day returns of cryptocurrency trading show the superior performance of the proposed model in terms of the identified risk patterns being informative with greater interpretability, enhancing the understanding of the systemic risks of a market and their causes, and making better risk management possible. This is joint work with Deyuan Li and Zhengjun Zhang.

This talk is a contributed talk at EVA 2021.