I got trolled too. Seriously though, I think it's more important to understand the nature of the 3rd order greeks than needing to calculate them. You should understand that your vega will change if you have an OTM option and you drift towards your strike. Similarly, you should realize that if you are short a deep OTM strike, then your gamma of gamma will be a big deal on a big move. But the actual risk numbers aren't intuitive and will rarely provide any meaningful pnl explanation or risk description. I like the theta idea. I hadn't thought of it before but it truely shows you your risk direction (except in certain skew situations) because it normalizes for expected gamma moves.
I don't see what the "big deal" is. 3rd order Greeks are just a measurement....they have no bearing on predicting direction of movement. They're in the same boat as RSI and all of the other technical indicators...they're just "Stats". Now, if you have a data mine of these stats, and can apply statistics to show a correlation with movement or volatility, then perhaps there is value in these.
Sorry, I must be misunderstanding you somewhere. There is nothing predictive about any Greeks and 3rd order greeks are no different. If i say that I am long delta, i am predicting nothing, I just know the response of my book to the movement in spot. Same way, if I say that I am long dVega/dSpot there is nothing predictive about it. Unlike technical indicators, the greeks are risk measures, not patterns. Greeks are there to aid your risk management and to understand the risk factors of the position/portfolio. You can, certainly, use 3rd order greeks to establish a volatility position - e.g. a short dVega/dSpot would give you exposure to the realization of the skew, and by definition, to vol of vol convexity.
One of the main reasons I want to look at 3rd order greeks is charm, color and speed. I have personally been burned on the change in gamma as an option closes in towards expiration, especially the last couple of days. AFAIK this "gamma decay", "delta decay" and "DGammaDspot" can only be measured through 3rd order greeks. DGammaDspot is of particular interest to me, as it helps predict changes in gamma as spot moves around, and this becomes important close to expiry with increasing gamma / delta decay. Accounting for DGammaDspot can thus give more accurate Pnl attributions and so on, etc... Great, thanks for this However when you change around the inputs I get a #NAME-error in the Gaussian CDF d1/d2 and Gaussian d1/d2 measures, which then result in #NAME-errors everywhere else on the sheet. EDIT: Nevermind, it's my stupid foreign language Excel that couldn't translate normdist properly. It works great.
But do these Greeks get you out of positions via rules ? IOW: Exit when gamma doubles, or Exit when Theta rises 50% ? If they aren't directly used in the decision making process, I don't "see" their value.
Looking at the option chain of spy does the option price different of individual price strike include the effect of volaility on delta? You may use the delta and gamma to compute the option price with price move up 1 point but how about volaility effect on delta for otm option
you can use your options software (ex: TOS) to see the what-if scenario what will happen to ITM,ATM, OTM options, when volatility changes.. for single-leg options, software can give you an idea how changes in options' IV affects your delta but not so much if you have multi-leg options as IV changes across the strike is not linear GL
Want to ask looking at the spy option chain delta from one price strike to another why is adding the gamma to delta does not give the delta of next delta? Isnt gamma the change of delta for one unit increase in price?
Erm, no. The delta changes due to the change in implied vol across the strikes too, so technically you should do something like (assuming some nomalized units, obviously): delta(k1) = delta(k0) + gamma(k0) * (k1 - k0) + dDelta/dVol * (iv(k1) - iv(k0)) + ... PS. Usually, when people calculate the delta they are making some sort of skew dynamics assumption. Simple assumption is that vol is fixed for a given strike, but that's not what you see in real life.
Why you have to use gamma ko × (k1-ko)? Isnt gamma the rate of change of delta? Should not delta be used rather k1-ko? And what are the ... behind?