You are using an out of date browser. It may not display this or other websites correctly.
You should upgrade or use an alternative browser.
You should upgrade or use an alternative browser.
gama-delta neutral
- Thread starter martin
- Start date
Johno, it seems that the "castle secrets" are being kept with the authors of these papers I reference. One guy moonlights for Goldman and writes about the theory, testing results, and conclusions rather than practical examples. There are formulas given for finding these hedging options but it would require someone familiar with higher math calculus. Those just reading the paper are like blind men trying to figure out the elephant. If you haven't read the paper then we can't even compare perspectives. By googling papers on "static hedging" of options you can see some of them, most of which deal with exotic derivatives. In general the premise of doing this static hedging violates several option principles, so it makes little sense on the surface and from a distance. So it is mystery what these academics are writing about.Quote from Johno:
When I referred to hedging off the greeks,I meant the individual risk pertaining to - Gamma, Theta, Vega etc.
I have no doubt that a quant can accurately replicate a portfolios' risk, but I don't believe that it can be done, in a practical sense, cost effectively as an ongoing approach. Consider just one greek - Theta - buying short dated options to hedge long term risk, bear in mind the objective of staying gamma-delta neutral and the volume of options required to achieve this. Volitility in this situation is usually higher than that of the portfolio being hedged - higher priced relatively speaking - whilst time decay is accelerating, remember on expiry a new hedge must be purchased.
Regarding your specific comments, I don't see how the IV of short-term options is usually any more expensive than longer-term IVs. When such occurs it would be a time skew and not the usual situation. You are correct that a new hedge would have to be purchased once a month for example, but compared with daily delta hedging that would be easy and cheap.
dmo, I agree with all of what you say in your excellent comment, which generally pertains to the effect of time on greek neutrality. However, IMO a more important reason that hedging with calendar spreads is so difficult is because the the variance in gamma with time.Quote from dmo:
But if there's little time remaining until expiration and the underlying goes to 105, you're going to become very short gammas, which will throw off your deltas. If the underlying moves to 100, you'll become very long gammas, also throwing off your deltas. You'll find it very difficult to remain gamma neutral, and therefore very difficult to remain delta neutral. You'll find yourself having to adjust often, with unpredictable results.
Also, if all your options have the same expiration, then gamma neutral pretty much equals theta neutral and vega neutral. But if you have options with different expirations (calendar spreads), that relationship breaks down completely. That's because vega and theta have an opposite relationship with time. With a lot of time remaining, vegas are high and thetas low. With little time remaining theta is high and vegas are low.
Let's forget about the paper on "static hedging" for a minute and look at the typical situation involved in trying to hedge shorter-term options with longer-term options such as when we try to achieve gamma-delta neutrality with a calendar spread. (and let's forget about vega now also). To use a common example, the gamma of a single shorter-term option can be equaled (hedged) by the gammas of two longer-term options. The delta imbalance in doing such would need to be neutralized by selling the shares/contracts. The downside of doing such would be the cost (Johno's concern) and the resulting sharp decrease in ROI. Thus it would seem that hedging calendar spreads would not be practical. Any comments on that?
Quote from mysticman:
Let's forget about the paper on "static hedging" for a minute and look at the typical situation involved in trying to hedge shorter-term options with longer-term options such as when we try to achieve gamma-delta neutrality with a calendar spread. (and let's forget about vega now also). To use a common example, the gamma of a single shorter-term option can be equaled (hedged) by the gammas of two longer-term options. The delta imbalance in doing such would need to be neutralized by selling the shares/contracts. The downside of doing such would be the cost (Johno's concern) and the resulting sharp decrease in ROI. Thus it would seem that hedging calendar spreads would not be practical. Any comments on that?
Mysticman - I come from the futures side, so I'll defer to you on the example using stock and options on stock.
In the futures world, there would be no added cost to hedge your deltas with the underlying. To the contrary, you have risk margining, so being delta neutral would actually reduce your costs. In any case, you can hold your margin in T-bills, so futures are considered to have zero cost of carry - that's the modification that Black made to create the Black 76 model for futures.
To me, the primary use of the type of calendar spread you describe would be to get long vega and remain neutral theta. For example, imagine a scenario where implied volatility is extremely low and I expect it to rise - I just don't know when. Complacency is high and I know that won't last forever, because panic follows complacency like night follows day. I don't want to buy straddles or strangles because I could lose money for months straight waiting for the move to come.
But if I buy 2 back month options and sell 1 front month option - and hedge my deltas one way or another - I'll be very long vegas and approximately theta neutral. And approximately gamma neutral. I can wait and wait. Then when a Bear hedge fund gets in trouble (to pick an example out of a hat) and everybody freaks out, I'll have my day in the sun.
Hi Mystic,
"I don't see how the IV of short-term options is usually any more expensive than longer-term IVs. When such occurs it would be a time skew and not the usual situation."
-----------------------------------------------------------------------------------
Quote from -The Elements Of Successful Trading p362-
"the implieds realy reflect the supply and demand situation of the option much more than as a predictor of the future volitility of the market"
Re time skew - time is most certainly consistent 1 day = 1 day = 1 day
the author then goes on to give practical examples of supply and demand in action. Implied volitility is different to Historical volitility!
The price skew that occures as options near expiry provides premium sellers with attractive profit opportunities.
My hybrid strategy includes bought and sold Verticals , Puts/Calls, Diagonal Calenders with directional as well as non-directional elements these are combined to provide best case RR.
Generally I don't pay to much attention to Quants theories as I usually find myself baffeled by the BS,which is counter productive as I'm trying to think clearly about specific trading problems, but thanks for the reference all the same.
Best Regards
Johno
"I don't see how the IV of short-term options is usually any more expensive than longer-term IVs. When such occurs it would be a time skew and not the usual situation."
-----------------------------------------------------------------------------------
Quote from -The Elements Of Successful Trading p362-
"the implieds realy reflect the supply and demand situation of the option much more than as a predictor of the future volitility of the market"
Re time skew - time is most certainly consistent 1 day = 1 day = 1 day
the author then goes on to give practical examples of supply and demand in action. Implied volitility is different to Historical volitility!
The price skew that occures as options near expiry provides premium sellers with attractive profit opportunities.
My hybrid strategy includes bought and sold Verticals , Puts/Calls, Diagonal Calenders with directional as well as non-directional elements these are combined to provide best case RR.
Generally I don't pay to much attention to Quants theories as I usually find myself baffeled by the BS,which is counter productive as I'm trying to think clearly about specific trading problems, but thanks for the reference all the same.
Best Regards
Johno
dmo -- a lot of my stuff is done with index futures also, so I know what you are talking about and agree with you. In fact, your post brings out the point that hedging the calendar gammas also neutralizes the thetas -- not desireable. All we would have left is the long vegas, so as you suggest this would be a good strategy when you want to wait out the low vols in the index futures markets.
Johno -- 1 day = 1 day has nothing to do with time skews. In fact I don't see how any of your quotes and references addresses the point that a time skew is not the usual state of affairs. Perhaps it is in the markets you trade?
Johno -- 1 day = 1 day has nothing to do with time skews. In fact I don't see how any of your quotes and references addresses the point that a time skew is not the usual state of affairs. Perhaps it is in the markets you trade?
Hello all,
I just read the entire post, but I am still a bit confused. Is the theta-premium selling (gamma-delta hedged) purely academic or people actually do this trade, profitably?
I just read the entire post, but I am still a bit confused. Is the theta-premium selling (gamma-delta hedged) purely academic or people actually do this trade, profitably?