Parameters for pricing option on EDF

Question

Ladies and Gents,

Im writing a quick routine to calculate implied vols for options on EUR$ futures with Bloomberg data. My question concerns the part where I have all my inputs and am ready to pass them to my implied vol function. Assume we have the following market quotes (example ticker: EDK1C 98.750 Comdty): Strike = 98.750 Underlying Spot = 99.7050 Option Price = 0.9550

When passing these to my function, do I convert the Underlying spot and strike to S = (100 - Underlying Spot)/100 and K = (100 - Strike)/100 respectively and use the market option price as is so our implied vol method is some function IV = f(S,K,Option Price,...) OR convert the option price to oP = 100 - (Option Price)*100 and leave the spot and strike such that our implied vol method is some function IV = f(Strike ,Underlying Spot,oP,...) ???

The latter has yielded a rational result but I would love some feedback.

Thanks.

Answer 1

Eurodollar options (and STIR options in general) are options on the RATE, so you have to transform the strike and underlying futures prices to yields as you mentioned (100-x)/100 AND you have to switch puts and calls. Also remember they are American exercise, although that's not very important these days. The volatilities are on the rates (log-relative returns). In Bloomberg, you have to use a lognormal model and no other models are provided. But many practitioners use other models and quote the 'normal' (basis point) vols instead of the Black-Scholes vols.

Answer 2

If it is option on future, you don't need the spot price at all. use the unaltered units for S and K, and you should get results for IV that make sense.

If you think they don't make sense, just plug your IV result back in and verify that you get the option price back correctly.

Answer 3

I'm not sure where the (100 - X)/100 pattern is coming from. Implied volatility is defined on top of the Black-Scholes value for an option, which I understood to multiply a decaying function of the dividend and interest-free rates with the current spot and strike...

In any case, I've always seen it expressed in terms of straight prices, no normalization necessary.