Dunno, but if the underlying really does exhibt GARCH effects, and the masses price it like a log-normal, you can take their money because you know when it's over- or under-priced? (in some expected sense, of course) GARCH effects can fatten tails, with a thinner "body" of the distribution of returns, relative to a Gaussian. Also looks a little like a Bayesian posterior if the underlying has hidden factors that evolve over time (and the system is practically non-ergodic as a result, I think) - might have to inject risk aversion to get that result, though.
To Steve: You know I copied your earlier reply (the fuck-one) so I might re-post it. Anyway, imagine this q in the TA forum: âCould you share a recent trade and explain how Bollinger Bands made it work?â Now Iâm not a TA fan but point is: no one would take offense at a question like this. Itâs just in the outskirts of mathland where they feel threatened when asked for any practical use for their scribbles: âWhaaat? You want to USE this? You mean... make MONEY off our precious theories? Desecrator! Pervert!â You see, with a doctorate in math I can actually follow at least the main thread of 20 pages of differential equations that lead to the conclusion that a contract should be priced 3.23 instead of 3.24 (not GARCH-specific, almost any academic piece on the option pricing subject). I taught math and started every new subject with an application of the theory. If there was none, I knew Iâd be in for a hard time keeping âm interested. And rightly so. So if you think your GARCH insights can improve the TRADING of âpeople of unknown expertise at a billboard site like ETâ (quote from your earlier post) then I am sincerely interested. And I mean that. This is probably not halfway enough but so be it. Back up to you now.
So with a doctorate in math you have no idea how to use GARCH I feel the word "Bullshit" developing in my throat. Tell you what "doctor" you take it from here. You have everything you need and I would only slow you down.. Good luck to you Steve Edit: I think your actions say quite a lot about your real interest. I have proven my willingness to help other many times over here at ET. All a person has to do is meet me halfway. You have another agenda, or you wouldnt take the time to copy replies and threaten to post them. I suggest you post it. Really I could give a fuck. Good luck to you.
Allright, but do we really need GARCH for that? Can't we just buy some deep OTM's and be done with it? (not my idea but Nassim Taleb's).
....... to improve my trading. Indeed, and I was hoping you could fill that gap. But just as I thought: deeply in love with the math, vastly disconnected from reality.
Look Steve, I didn't start this thread but aPismoClam did and I don't want to be in the way of you helping him out. I'll sit in the back row and keep quiet.
Do what you want sir, I am not the ET police I do have one more comment regarding the practical applications of GARCH First, in my opinion once a person understands how to use GARCH and can fit a model to existing returns, it is then possible to forecast volatility. This can be done with a number of stat packages. The one I have used most often is S+Plus. The forecast you obtain with the package is referred to by skilled users as the term structure of volatility and it reflects a non-gaussian distribution. I assume now that I have lost some folks. What I am saying is that once you know how to manipulate the model (GARCH in a statistical package) you can get a more accurate reading of what the true value of an instrument (like an option) is, not only in the present but in the future. Presumably if you see an option whose price is way out of line with your model, it is because somebody either A.) knows something you don't or B.) They have made a mistake in pricing the option For anyone interested S+Garch Toolkit is a data analysis product from MathSoft in Seattle WA. Using that toolkit you can check Derivatives pricing, Volatility forecasts, and Value-at-Risk management for a portfolio of instruments. The toolkit is an object oriented software implementation, so it can be learned pretty quickly.
Garch is also implemented in R (free). Besides their application in option pricing, Garch and other volatility models allow to estimate risk of trading / investing strategies. For example, your developed 2 trading systems: 1) Annual Profit: 50%, Max DD: 10% 2) Annual Profit: 20%, Max DD: 2% Even though (1) has a higher expected profit, you might consider trading (2) with leverage and obtain a similar return with less risk, or a larger return with similar risk. The confidence you'll have in making this decision will depend a lot on how precise your return and risk estimates are. Garch might help you to feel more confident about the risk estimate.
FWIW, there have been several studies on the benefits of using the various GARCH models for forecasting volatility on broad index products. The consensus was that these models were no better than the predictive qualities provided by ATM implied volatilities. A quick Google should throw up the papers I am referring to. However, it is always useful to have an additional tool especially when you feel that IV may be slightly misleading due to extreme fear or greed. 2 cents. MoMoney.
welp, glad you guys enjoyed this thread. back to the textbooks and Wilmott forum for me. Hey, moron, what's your issue with Jesus Christ? Why do you spout idiotic things about Him at every opportunity? You know, I did that for a while when I was twelve years old and used to write short stories. I daresay, the time may come when you realize how juvenile (and curious) your obsession is.