CBOE Variance Futures

[cdcaveman]

Quote from cdcaveman:

one day i'm going to need a two way conversation with someone that can really explain it to me... i appreciate your help.. i' m just still in the kindergarten when it comes to all this

if i can get it.. ANYONE can! haha

 

[sle]

Quote from heech:
The only way to get this number is to download the calculator from the CFE website. You can plug in hypothetical values for implied + vega notional, and get back precise numbers in terms of futures + price.

Actually, it's simple enough - in a regular variance swap, variance units relate to vega notional as
Var Units = Vega Notional / ( 2 * Variance Strike )
In futures, to keep vega consistent across time as in a seasoned variance swap, you need to rescale it by the ratio of total effective days over days left:
Futures Units = Var Units * (N effective - 1)/(N remaining - 1)

Quote from heech:

But again, unless I'm wrong about commissions on this thing, then this seems dead on the vine to me. (Vision is studying it for me.... We will see if they can give me penny commissions or something to make this worthwhile.)

Don't know about commissions, however the exchange fees (which, usually are more or less 50% of the total vig in futures) are $8 per 1k of vega notional for a customer. In terms of expected volatility you should expect that your position (Feb, is it?) has a volatility equal to about 1.5 times of the Jan VIX futures (vol of vol of implied plus the realization cushion). 1 vol a day is totally reasonable (say vix of 16 times 100% annual vol gives you about 1 vol a day) and you should give it some space for convexity - a 2 vol day is actually more then 2.

 

[heech]

Yea, there definitely needs to be a "fix" here for commissions... Vision is looking at doing something custom for me, but based on my standard setup:

- Vision charged me $2.40/side in fees yesterday,
- FC Stone charged me $1.52/side...

So, for 1k vega notional, that's $100-200 in fees. Basic exchange fees should only be $8 (as you mentioned above). So... yea. Time to have a discussion with my FCMs.

 

[sle]

Yeah, that's a lot. I have the historical variance data for you, gents, don't know if I should post it or I should send it via email upon request.

 

[Jgills]

Quote from sle:

Yeah, that's a lot. I have the historical variance data for you, gents, don't know if I should post it or I should send it via email upon request.

i like the email upon request.. until you get like 1500 emails..

i request por favor.

 

[sle]

Would you rather get constant maturity or fixed maturity data? That is, for each date either something like 1m, 2m, etc or Jan12, Dec12?

 

[2rosy]

Quote from sle:

Yeah, that's a lot. I have the historical variance data for you, gents, don't know if I should post it or I should send it via email upon request.

you have historical variance data or historical variance swap quotes?

 

[sle]

Quote from 2rosy:
you have historical variance data or historical variance swap quotes?

OTC variance swap levels - I would hope that people can calculate historical variance on their own using data from yahoo

All I am asking is since I have them stored and parametrized, i can output CM form or just dump the original quotes. Original quotes are a bit harder to deal with but on the other hand, you can lose some information about liquidity and/or calendar effects if you look at the CM form only.

 

[Jgills]

Quote from sle:

OTC variance swap levels - I would hope that people can calculate historical variance on their own using data from yahoo

All I am asking is since I have them stored and parametrized, i can output CM form or just dump the original quotes. Original quotes are a bit harder to deal with but on the other hand, you can lose some information about liquidity and/or calendar effects if you look at the CM form only.

if you're going to offer me the choice... can i get both?

if not, i would prefer fixed maturity, original quotes

 

[sle]

Quote from Jgills:

if you're going to offer me the choice... can i get both?

No. I want a consensus because I have a limited amount of time I can spend on this - right now at a meeting so can bs about it, but not do actual work.

CM is better for analysis, but it comes from an SPX vol surface plus variance offsets which was used for market making so it's not a perfect "market" data.

Quotes are just a BBG chat scrape with futures ties. Beware that there will be days without any quotes and also the quotes will not be tied to the same spot as the parametrized data would be, you would have to delta-adjust it on your own.