How to Validate Model-Based Derivative Prices

Summary

The document raises the problem of checking whether prices produced by models such as Black–Scholes or Heston are accurate. It distinguishes a priori validation, which includes choosing a model and fitting it to market data, from a posteriori validation after the model has produced a price. Its central question is what benchmark can meaningfully assess that output.

The text is framed as a question rather than a complete answer: it offers no validation procedure, empirical comparison, or findings. It is useful as an introduction to the distinction between selecting and calibrating a pricing model and evaluating its resulting prices. Any practical assessment would depend on the instrument, available market observations, and the purpose of the model, none of which are developed here.

Key ideas

• Model price validation can be considered before or after a model produces prices.
• A priori checks include model selection and fitting to market data.
• A posteriori validation requires a suitable benchmark for comparison.
• The document poses the validation problem but does not propose or test a specific benchmark.

Tags

# How are quants able to verify whether their calculated prices are any good


# How are quants able to verify whether their calculated prices are any good












This question is related to the discussion on Model Validation Criteria However it appeard to be very high level to me and I would like to go more into detail.

Not working at a pricing desk the following has always puzzled me: How are quants able to verify whether their calculated prices are any good? (by "calculated prices" I mean the ones output by some model - e.g. B&S, Heston )

One could perhaps distingusih the two:

- A priori

- A posteriori

I would assume a priory mainly consists of model choice (fitting the market data etc.) - Thus the A priori part is related to this question.

Alas, I am quite clueless what an a posteriori check would look like. What would be the benchmark one would check against ?

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.