Why Risk Management Uses Benchmark Models to Adjust a Heston Volatility Surface

Summary

The document raises a model-risk question: when a trading desk uses the Heston stochastic-volatility model to generate a volatility surface, why might risk management use other models to produce adjustments? It names interest-rate mean reversion and other local-volatility/stochastic-volatility approaches as examples, and asks how their role relates to the desk’s primary model.

The prompt points toward a practical distinction between a model used to fit or produce market prices and models used to test risks that depend on assumptions or dynamics the primary model may not capture. However, the document contains no explanation, comparison, calibration details, adjustment procedure, or evidence about the specific models named. It does not establish which adjustments are appropriate or how they should be computed. Readers can use it to identify the topic of model benchmarking, but need further material to learn a method for applying it.

Key ideas

• The document asks why risk teams may adjust a Heston volatility surface using other models.
• It gives interest-rate mean reversion and other local-volatility/stochastic-volatility models as examples.
• The question concerns how a desk’s production model relates to risk-management benchmarks.
• No adjustment method, model comparison, or evidence is supplied.

Tags

# Benchmarking of models


# Benchmarking of models












Why do we need benchmarking adjustmenets for a model? Suppose, a trading desk is using Heston model for generating vol surface, then why do risk management uses various other models like IR Mean Reversion, other LVSV models etc to generate adjustments to be made to the Heston Model?

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.