Publication:
Measuring asset market linkages: nonlinear dependence and tail risk

Loading...
Thumbnail Image
Identifiers
Publication date
2019-08-25
Defense date
Advisors
Tutors
Journal Title
Journal ISSN
Volume Title
Publisher
Taylor & Francis
Impact
Google Scholar
Export
Research Projects
Organizational Units
Journal Issue
Abstract
Traditional measures of dependence in time series are based on correlations or periodograms. These are adequate in many circumstances but, in others, especially when trying to assess market linkages and tail risk during abnormal times (e.g., financial contagion), they might be inappropriate. In particular, popular tail dependence measures based on exceedance correlations and marginal expected shortfall (MES) have large variances and also contain limited information on tail risk. Motivated by these limitations, we introduce the (tail-restricted) integrated regression function, and we show how it characterizes conditional dependence and persistence. We propose simple estimates for these measures and establish their asymptotic properties. We employ the proposed methods to analyze the dependence structure of some of the major international stock market indices before, during, and after the 2007-2009 financial crisis. Monte Carlo simulations and the application show that our new measures are more reliable and accurate than competing methods based on MES or exceedance correlations for testing tail dependence. Supplementary materials for this article are available online.
Description
Keywords
Expected shortfall, Market crashes, Tail risk, Nonlinear dependence
Bibliographic citation
Escanciano, J. C., & Hualde, J. (2019). Measuring Asset Market Linkages: Nonlinear Dependence and Tail Risk. Journal of Business & Economic Statistics, 39 (2), pp. 453-465.