FindTheTail - Extreme Value Theory - Matlab
FindTheTail automatically determines the optimal threshold for the tail of an unknown parent distribution. A distribution´s tail is modelled via a Generalized Pareto Distribution (GPD) to assess extreme negative tail risks. Risks are quantified by the Value at Risk and Expected Shortfall (cVaR) for different confidence levels.
The FindTheTail function and its documentation can be found in the Live Script.
Cite as: Pascal Bruhn (2022). FindTheTail - Extreme Value Theory (https://github.com/PascalBruhn/FindTheTail/releases/tag/v1.1.1), GitHub.
References:
- Hoffmann I, Börner CJ. Body and tail: An automated tail-detecting procedure. J Risk [Internet]. 2021; Available from: https://www.risk.net/journal-of-risk/7733836/body-and-tail-an-automated-tail-detecting-procedure
- Hoffmann I, Börner CJ. The risk function of the goodness-of-fit tests for tail models. Stat Pap. 2021 Aug;62(4):1853–69.
- Hoffmann I, Börner CJ. Tail models and the statistical limit of accuracy in risk assessment. J Risk Finance. 2020 Jun 27;21(3):201–16.
Practical examples:
- Börner, Christoph J. and Hoffmann, Ingo and Krettek, Jonas and Kuerzinger, Lars and Schmitz, Tim, On the Return Distributions of a Basket of Cryptocurrencies and Subsequent Implications (May 14, 2021). Available at SSRN: https://ssrn.com/abstract=3851563 or http://dx.doi.org/10.2139/ssrn.3851563
Contact:
Pascal Bruhn, pascal.bruhn@financial-research.de
Created in cooperation with Christoph J. Börner and Ingo Hoffmann (Heinrich Heine University Düsseldorf).