Conditional Scenario Generation with a GVAR Model

Stress-testing;Conditional Scenario Generation;High-dimensional Data;GVAR

Wang, Xinghao

University of Waterloo

关键词: Stress-testing;    Conditional Scenario Generation;    High-dimensional Data;    GVAR;   

Others  :  https://uwspace.uwaterloo.ca/bitstream/10012/11108/5/Wang_Xinghao.pdf

瑞士|英语

来源: UWSPACE Waterloo Institutional Repository

【 摘 要 】

The stress-testing method formed an integral part of the practice of risk management. However,the underlying models for scenarios generation have not been much studied so far. Inpast practice, the users typically did not model risk factors for portfolios of moderate sizeendogenously due to the presence of ;;curse of dimensionality;; problem. Moreover, it is almostimpossible to impose the expert views for a future outcome of macroeconomy on thescenario generator without making ad-hoc adjustments.In this thesis we propose a GVAR-based framework which allows an efficient simulation ofrisk factors for a complex multi-currency portfolio of various classes of assets conditioning oneconomic scenarios. Given reasonable sets of economic forecasts, the GVAR model anticipatesthe trend and codependency of the future path of portfolio risk factors and supports theproduction of meaningful results from risk analytics.