Kenneth ChengMathematical modeling

Editorial project overview

Heavy-Tailed Financial Risk Stress Test

How much tail risk can Gaussian forecasts hide when volatility clusters and losses include heavy tails and jumps?

Reproducible studyUpdated 2 Aug 2026
Technical figure from Heavy-Tailed Financial Risk Stress Test
A reviewed research figure synchronized from the private technical workspace. The original aspect ratio and labels are preserved.

Editorial overview

Read How Much Tail Risk Does a Gaussian Model Hide? for the stress process, diagnostics, VaR and ES forecasts, backtests, calibration, tail-risk surface, uncertainty, sharpness frontier, and nine approved figures.

The public page interprets reviewed evidence without exposing private code, fitted arrays, notebooks, or intermediate calculations.

Key findings

  • Filtered historical simulation is closest to the nominal 1% exception rate at 0.00972.
  • The fitted Student-t GARCH degrees of freedom is 4.2685, indicating materially heavy innovations.

Limitations

  • The returns are synthetic and the results are not investment advice.
  • Multivariate dependence, liquidity, transaction costs, and structural breaks are outside scope.

Technical record

Detailed source, calculations, generated figures, and reproduction instructions are maintained in a private technical workspace. Public articles contain only manually reviewed interpretation and approved figures.

Version history

2026-08-02 — Curated overview reviewed against repository evidence.