8
$\begingroup$

The CME-traded Eurodollar futures option is an American option.

What is the industry standard pricing model for this product?

Does the industry practice to treat CME-traded Eurodollar futures option as European and use Black-Scholes model to price it (due to daily margin)?

$\endgroup$
3
  • 2
    $\begingroup$ @Emma - while your link is about Eurodollar options - it does not provide or hint at an answer regarding the pricing model question whatsoever. $\endgroup$ Commented Feb 22, 2019 at 7:41
  • 3
    $\begingroup$ I stand by the word "whatsoever". The passage you cite merely states that the options are American style which is something that the OP already knows and says in the question. This question is about how they are priced in practice and whether approximating the true value of the American price using a European pricer is sufficient or not. The PDF contains no insights regarding this. $\endgroup$ Commented Feb 22, 2019 at 9:21
  • 3
    $\begingroup$ I think you fundamentally misunderstand the question. It is about option pricing models and in particular whether or not to explicitly incorporate early exercise in the option pricing. What you just cited is what drives the price of the underlying futures. Similar issue as in your deleted answer to quant.stackexchange.com/questions/43962. You can edit your comments by the way - no need to post three of them. $\endgroup$ Commented Feb 22, 2019 at 9:26

2 Answers 2

9
+50
$\begingroup$

Having traded these options for a number of years I have some insight. It’s my belief that those that make a living specifically out of these options do have tree-style models that take into account early exercise. On the other hand , those that have occasional use of these options (such as interest rate derivatives dealers who might use them to hedge otc derivatives) mostly assume that the options are European, because implementing the full American model is cumbersome within a multi- product portfolio.

It is possible to show that the error committed in assuming the option is European is not very large. Consider the case of a call option on the Mar2020 contract (approx 1year to expiration) , struck at 9450, with the contract trading at 9750. This contract is deep in the money (actually, 300bp In the money). Left unexercised, it is worth 300bp, discounted for one year. However , upon exercise , you get delivered the underlying futures contract, together with 300bp of immediate variation margin. This situation is worth 300bp , not discounted. Hence , it is my experience that the modeling error only shows up for long dated, deep in the money options , and then only when interest rates are high enough to make a discounting difference.

I hope that helps.

$\endgroup$
2
  • $\begingroup$ How would that tie in with the fact that for these options the premium is usually not paid upfront but rather margined daily (just like for the underlying futures contract itself)? That should render early exercise to never be optimal (see e.g. Oviedo 2006 paper), regardless of maturity and/or rate level? $\endgroup$
    – KevinT
    Commented Apr 11 at 9:41
  • $\begingroup$ Hi, option premium is actually not margined like a futures contract. It is more like an otc derivative whereby you pay/receive overnight interest on any variation margin. $\endgroup$
    – dm63
    Commented Apr 12 at 10:01
2
$\begingroup$

To add to @DM63's answer, as a secondary characterisitc, vol may also matter in deciding the european approximation impact. As vol goes to 0, you want to exercise as soon as possible, because the underlying future rate becomes a constant (a martingale with no vol). As you'll receive the same payoff at at date, better to get it earlier (if rates are.positive).

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.