×

Simultaneously capturing multiple dependence features in bank risk integration: a mixture copula framework. (English) Zbl 1454.91334

Lee, Cheng Few (ed.) et al., Handbook of financial econometrics, mathematics, statistics, and machine learning. Volume 2. Hackensack, NJ: World Scientific. 1485-1518 (2021).
Summary: This chapter proposes a mixture copula framework for integration of different types of bank risks, which is able to capture comprehensively the nonlinearity, tail dependence, tail asymmetry and structure asymmetry of bank risk dependence. We analyze why mixture copula is well-suited for bank risk integration, discuss how to construct a proper mixture copula and present detailed steps for using mixture copula. In the empirical analysis, the proposed framework is employed to model the dependence structure between credit risk, market risk and operational risk of Chinese banks. The comparisons with seven other major approaches provide strong evidence of the effectiveness of the constructed mixture copulas and help to uncover several important pitfalls and misunderstandings in risk dependence modeling.
For the entire collection see [Zbl 1446.91002].

MSC:

91G40 Credit risk
62P05 Applications of statistics to actuarial sciences and financial mathematics
62H05 Characterization and structure theory for multivariate probability distributions; copulas
Full Text: DOI