Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/172002 
Year of Publication: 
2017
Series/Report no.: 
Bundesbank Discussion Paper No. 34/2017
Publisher: 
Deutsche Bundesbank, Frankfurt a. M.
Abstract: 
The severity function approach (abbreviated SFA) is a method of selecting adverse scenarios from a multivariate density. It requires the scenario user (e.g. an agency that runs banking sector stress tests) to specify a "severity function", which maps candidate scenarios into a scalar severity metric. The higher the value of this metric, the more harmful a scenario is. In selecting a scenario the SFA proceeds as follows: First, it isolates a set of equally severe scenario candidates. This set is determined by the condition that more severe scenarios only occur with some user-specified probability. Second, from this set it selects the candidate with the highest probability density, i.e. the most plausible scenario. The approach hence operationalizes the mantra that "scenarios should be severe yet plausible".
Subjects: 
Stress Testing
Conditional Forecasting
Density Forecasting
Time series
Bayesian VAR
Simulation
JEL: 
C11
C32
C53
C61
G01
G32
ISBN: 
978-3-95729-409-8
Document Type: 
Working Paper

Files in This Item:
File
Size
692.93 kB





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.