Abstract
Without much ado, Chap. 4 outlined a relatively straightforward historical VaR model. In the bank I worked at, such a model proved to work reliably right through the 2008 financial crisis and its aftermath. Many a model aspect, however, could be tuned or tweaked or altered, and this chapter zooms in on some of those model choices. But how to weigh these features, how to choose between model options? Let me give you my personal take on this.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
Such overrides are not uncommon. If, for example, separate models yield probabilities that should sum up to 1, the probabilities can be normalized to enforce an exact match or identity to 1. If one model fails and yields, say, 230%, such a step, if thoughtlessly implemented, may cover up and obfuscate a breakdown more apparent otherwise.
- 2.
Just as nobody got ever fired for buying IBM, no vola model was ever rejected for relying on GARCH.
- 3.
The lowest naturally occurring local volas “in the wild” remain unaffected by this floor. The standard deviation of 19 zero returns combined with a return of one one-hundredth of one basis point is—with about 10−7—already much larger than this purely technical floor.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer International Publishing AG
About this chapter
Cite this chapter
Auer, M. (2018). Model Choices. In: Hands-On Value-at-Risk and Expected Shortfall. Management for Professionals. Springer, Cham. https://doi.org/10.1007/978-3-319-72320-4_9
Download citation
DOI: https://doi.org/10.1007/978-3-319-72320-4_9
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-72319-8
Online ISBN: 978-3-319-72320-4
eBook Packages: Economics and FinanceEconomics and Finance (R0)