Well the question "which volatility do you use to manage an options position in which each strike trades at a different IV" is an excellent one with no widely-agreed-upon answer and many suggested ways of handling it. You can divide the solutions into two broad categories - calculate your deltas (and other greeks) using a different volatility for each strike, or simply use 1 volatility across the board (generally the ATM volatility). If you decide to use different vols for each strike, then which vols to use? The really professional programs like Pro Opticus by Prime Analytics have several ways of doing it that are frankly beyond my expertise - interpolation using curve fitting, cubic spline, and others whose names I don't recall. Being a simple sort, I personally prefer to use the simplest solution, which is to use the ATM vol across the board for calculating my entire position at every strike. There are lots of reasons, but I guess it comes down to the fact that it works for me. So the answer to your question is that in my example I use the ATM volatility to calculate the delta of every strike. How often do you even up your deltas and lock in a profit? That too is an eternal question with no easy answer. It's the same question as "I'm in a winning trade, when do I take profits?"
If I understand this correctly this is like a 'straddle' on vega? You risk is the path dependency on the underlying and some theta exposure? Just to clarify for myself, you wouldn't ratio spread the calls to account for the skew because your d-vega exposure would be too great on the call side? ex. Buy (1) 1100 P @ 50 vol, Short (2) 1300 C @ 25 vol, delta neutral.
Going back to the original example knocks - this was an example of what I WOULD do if I could buy the 1100 put and sell the 1300 call at the same volatility. I would never in a million years pay 50% vol for the 1100 put and sell the 1300 call for 25% vol. When I put it on I would do it vega-neutral (and of course delta neutral). Since in my example the futures are right between the two strikes when I put it on, I don't need to ratio it. If the futures were a little higher or a little lower, I would probably do a slight ratio to get vega neutral (and of course get delta neutral with the underlying).
Basically, I keep the same implied volatility while all the position lasts. That is the implied volatility the strike traded, the day traded. This is the very simple way I found to compute my pos. But I change my deltas. Because of the skew, I add the variation of the volty with respect to the spot to my previous delta (I interpolate the sensitivity by the skew between strike as proxy). My modified gamma is compute by the up and down range I've choosen for my delta. I finally add a little vega (something like that) to the extrem values. My daily PnL is something like (0,5*modified gamma*daily spot variation*dailyspot variation)-theta for each long positions, the opposite for the short ones. Course my traded hedge is finally a global one.
....quoting dmo.... The other side? You mean if you start out delta neutral and short gammas? I would just call that being short premium. In that case you'd want to lose the scalping bit, because negative gammas means every time the underlying moves your deltas move against you and you lose money. So every time you even up your gammas you're locking in a loss. ........................ good point dmo, Since I've been known to trade IC's more than once (please don't lecture me about the risk--I understand the risk/reward ratio and watch them quite closely), I know a little about being delta neutral and short gammas--- I'm usually close to delta neutral at the start and after any adjustments, and always gamma negative as well as looking to profit from time decay. The scalping issue is one of the chief drawbacks of the whole IC scenario-- in every case, any major movement either requires nerves of steel coupled with a desperate hope for a bounce back or an adjustment that costs you money (negative gamma scalping if you will). The only exceptions are when the theta decay has been working for you for a week or two. Then, you may be able to adjust and take profits on both sides. That is what I usually hope for--sometime during the month when the market simply moves back and forth a bit aimlessly so that the premiums will shrink. If that does not happen, the adjustments can become somewhat painful, but I would rather adjust than see my shorts go into the money.
I was going to object to this entire thread, but I figured I would just wait a few days to disprove just this not only in theory, but in practice. I strongly suggest that you look at the VIX today and look at SPX or ES. nitro
Hi nitro Please don't. If you read some posts of Dmo you know that he got a problem with words like "always, never, ever, all, every, none..." But he really got a gift how to explain in plain english the rules and tricks to understand option variations. That is the most important here on ET. I don't know if he tries to correct himself but I know that he made some concepts reachable for newbies.
Nitro, I don't think I ever said that if someone watches day in and day out he will never see an exception. But the exception does not disprove the rule - quite the contrary. It's true that this morning's correlation between the SPX and the VIX was the most unusual I've ever seen - I noticed it at the time and even PM-ed a friend of mine about it. I believe that nervousness about LEH and Ike took precedence this morning, goosed up the VIX, and thus distorted the normal relationship for a few hours. But even today, from about 11:40 central time until the close, the normal negative correlation reasserted itself (I drew in the 11:40 time bar). I'm posting today's one-minute chart as proof. This is about the umpteenth chart I've posted on this - so far I've yet to see a chart posted in rebuttal.
I second that motion, he has given out more info for free than you would get in a $2,0000 seminar, along with riskarb option coach and a few others. By the way if you you write an ebook for this type of info I would gladly buy it.