VIX Options Pricing

I'm looking into the pricing of VIX options and have a couple of questions.

Which pricing model do you use?

This paper (2009) states that no model is correct across all strikes.

https://msb040.msb.edu/faculty/bodu...Empirical Evidence Beyond Simulation_2009.pdf

Secondly, in using VIX Futures price as the price of the underlying, do you use the Futures expiration that corresponds to the Options expiration (eg VIX November Futures for VIX November Options) or do you use 1 month later for the futures (VIX December Futures for VIX November Options)?

This article states "....so a shortcut that seems to be working so far is to price each different month of $VIX options using that month’s futures price as the underlying price – not the price of $VIX itself. "

http://www.optionstrategist.com/blog/2011/08/basics-vix-options

While this one states " In other words, prices for VIX options expiring in the month of May will most closely reflect the level of the June futures contract while prices for VIX options expiring in August will most closely reflect the level of the September futures contract."

http://www.theoptionsinsider.com/tradingtechnology/?p=1185&qcABC=1


P.S. If anyone is interested in a general view of VIX Futures and Options, I found this useful.

http://www.futuresmag.com/2010/08/18/understanding-vix-futures-and-options?t=options

 

(a) Black76 is good enough for 90% of things you'd need
(b) True. That is also true for options on any other asset, e.g. single stocks, S&P index, IR swaps or credit indices.
(c) Dec options should use Dec futures prices as an underlying
(d) This is either complete bullshit or I am misreading it
(e) Be weary of what you find on the internet, you will grow hair on your palms

 

Thanks.

Re. (d), no you are not misreading it. It did not make sense but the guy who posted the article is "Randy Frederick (is) Director of Derivatives at the Schwab Center for Financial Research", so I figured I'd better check.

 

I have a similar question. I trade with IB. When using the risk navigator it uses the spot vix instead of the futures to determine the greeks.


Any rules of thumbs or ideas on how to correct this?

I know the model is a simple one but having the input off so much can create some big differences.

Thanks

 

wow haha don't use that !

 

You can use Model Navigator to change the underlying to the appropriate futures contract and it will update the greeks accordingly. When I use it though it changes the underlying for all the VIX option expirations to that specific futures contract, not just the expiration I want (although presumably there is a way to change the underlying for each option expiration individually and save that preference, I just have not figured out how to do it but admittedly have not spent much time with it).

 

Look at this video.
http://www.youtube.com/watch?v=D7bOYMxEqVc.

 

Actually looking at the VIX option deltas for Nov 19 '13, vs the future for Nov they don't check up. For instance the 17 call contract has deltas of 0.52 implying the future should be at ~17.2 but it's at 16.62 now...maybe there are other factors I'm missing. I may be way off

The only explanation I can think of is contango being worse in VIX options as they have contango on top of extrinsic value, but not sure why it would manifest like that.

 

Well, the explanation is much simpler - 50-delta is not nessesarily ATM forward. The difference between ATM forward and the 50-delta strike is exp(-0.5*time*vol^2), so the higher is the implied volatility, the more 50-delta strike would diverge from the ATM forward.

 

Does anybody know what algorithm IB uses to calculate their model greeks ? The TWS shows a delta and a model delta. But how do they calculate this ??? Black&Scholes ? Any other GARCH, etc....