Re: opt_bs bug?

[Hal <hal ashburner info>]

On 03/30/2010 12:31 AM, Scott Ballantyne wrote:
> Looks like the opt_bs function is actually the opt_b function. It
> calculates the Black model for European options on futures contracts,
> instead of the original Black-Scholes for European options on
> non-dividend paying stocks.
>
> Black formula for calls:
>
> C = Ue^(-rt)N(h) - Ee^(-rt)N(h-vt^0.5)
>
> Black-Scholes:
>
> C = UN(h) - Ee^(-rt)N(h-vt^0.5)
>
> and so on through the opt_bs_delta, etc.
>
> It would be nice to have this too, when someone has the time.
>
> Until then, may I suggest the documentation be changed to reflect
> this?
>
> Thanks again for all the wonderful work on Gnumeric. I couldn't live
> without it.
>    

Hi Scott,
Cost of carry is the optional last parameter to the function so in my 
understanding that makes it capable of calculating either options on 
assets with no dividends (such as options on futures), by setting this 
to be zero. Or options on dividend paying assets such as common stocks.

I'm not sure how many people have actually used these in anger since I 
submitted them. It does seem that there are a number of conflicting 
ideas about definitions, names of parameters and so on, which, even if 
I'm all wrong, mostly don't agree with each other.
So I'm happy to make changes if:

There's an actual clear practical use for it that is actually going to 
be used by somebody in anger.
And it's not just my textbook says it's called X where yours says Y.
Or
The actual gnumeric maintainers think the change is a good idea, I'll 
defer to their judgement.

Thanks for the kind words,
Hal