Software Resources for Testing Dragon-King Crash Models

Summary

This exchange concerns tools for investigating the Dragon-King hypothesis, which proposes that some financial crises and crashes may be predictable rather than behaving like ordinary extreme events. The question asks for software or code to test that idea. The responses point to Didier Sornette's Financial Crisis Observatory, a separately developed Java-based tool, and a Python package for calibrating a model associated with the approach.

The document is primarily a resource list, not a tutorial. It does not explain the statistical tests, calibration choices, data requirements, or how to assess forecast quality, and it reports no predictive results. Finding or running a package would therefore be only a starting point: researchers would still need to understand the model assumptions, handle fitting and validation carefully, and distinguish retrospective pattern detection from useful out-of-sample forecasts. The document offers no comparison of the listed tools or guidance on whether their implementations remain maintained, so it cannot establish which option is most suitable.

Key ideas

• The Dragon-King hypothesis proposes that some extreme financial events may have identifiable predictive structure.
• The exchange names tools for exploring the hypothesis, including a crisis observatory and model-calibration software.
• The listed resources do not substitute for understanding model assumptions and statistical validation.
• No software comparison or evidence of out-of-sample predictive performance is provided.

Tags

# Tools/R code for predicting Dragon-Kings


# Tools/R code for predicting Dragon-Kings












The theory of the so called Dragon-Kings, esp. by Didier Sornette (ETH Zürich), basically states that financial crises and crashes are predictable (contrary to the theory of black swans).

The following paper gives an overview (see esp. section 5, p. 20f. for predictability): Dragon-Kings: Mechanisms, Statistical Methods and Empirical Evidence by Didier Sornette, Guy Ouillon

My question Do you know any software, tools, Excel sheets and/or preferably R code/packages with which the predictability of Dragon-Kings can be tested?

## Answer by Solar Anamnesis (score 6, accepted)

https://quant.stackexchange.com/a/16950

My favorite tool is Sornette's own Finanical Crisis Observatory: http://tasmania.ethz.ch/pubfco/fco.html

If you are interested, I have developed my own tool in Java and JavaCL which can be found here: https://thebubbleindex.codeplex.com/

Update: Code moved to github: https://github.com/thebubbleindex/thebubbleindex

## Answer by joshwa (score 1)

https://quant.stackexchange.com/a/66567

Here is a Python package that calibrates the model: https://github.com/Boulder-Investment-Technologies/lppls

Shown in full with attribution under the source's licence. Licence: CC BY-SA 4.0 (Stack Exchange)

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.