Why Early-Exercise Rainbow Options Challenge Monte Carlo Pricing

Summary

The document frames a derivatives-pricing question about which traded instruments are hardest or infeasible to value. Its concrete example is a best-of or other rainbow option with early exercise. A rainbow payoff depends on multiple underlying assets, making simulation a natural candidate for valuation, while the early-exercise feature introduces a decision about whether to exercise before expiry. The question highlights the tension between those features and asks whether such a contract can be priced and how.

No answer, pricing algorithm, market example, or evidence is supplied, so the document does not establish that this option is impossible to value or that Monte Carlo methods cannot be adapted. It is useful chiefly as a problem statement about the interaction of multiple underlyings and early exercise. Any practical valuation would require further specification of the payoff, exercise rules, market model, and numerical method; none of those details are provided here.

Key ideas

• A best-of option is a rainbow derivative whose payoff depends on multiple underlyings.
• Early exercise adds an exercise decision over the life of the contract.
• The document raises, but does not resolve, how to combine multi-asset simulation with early-exercise valuation.
• It provides no pricing method, traded example, or evidence that the contract is impossible to value.

Tags

# What are the most difficult/computationally expensive/infeasible derivatives to price?


# What are the most difficult/computationally expensive/infeasible derivatives to price?












I'm not sure if this question has a concrete answer or if it's more of a fun game, but I suppose the question that does have a concrete answer is what's the most difficult instrument to value that has actually been traded OTC?

For the game part of it, what is the most difficult thinkable instrument to value, and if there are several completely impossible then what is the "least complex" but still impossible to value theoretical instrument?

As a start I was considering something like a Best-Of option (or any rainbow option) (typically suited for Monte Carlo valuation) but allowing for early exercise (typically not suited for Monte Carlo valuation). Can something like this be valued and how would you do it?

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.