Most traders i know found GARCH to be of little use for trading. In general, minor improvements in quality of prediction are less important then predicting exposure to shocks and their impact. So, your ability to incorporate information that is external to the price series (e.g. macro, events, PE, peers etc) is going to improve your ability to trade vol much more then using the latest version of GARCH.
A lot of time the underlying will slow down in trading going into an event essentially making your garch forecast even worse... I like the saying about averages... Would you cross a lake that was on average 4 ft deep...
This makes sense. However what most traders mistake is that there aren't boundless opportunities that will actual formulate for each individual that will have the risk reward worth taking. There are many opportunities in hindsight Vol trading is just another means to bet, not sure it's any better than underlying, it just seems to have a better leverage function than underlying than anything else if your " somewhat" right in regards to price and or vol. Because its a bit complex IE Greeks there seems to be a lot of subdifuge about how options are used to bet Right or wrong?
Since I know you are a Livevol user; be aware that their proprietary indicators, such as IV30, is a weighted, hypothetical measure of the volatility of a rolling 30-day option. This differs from the volatility of an option of a discrete expiration cycle. The measure of a stock's historical volatility, as mentioned in previous posts, will often include events, such as earnings, that will alter your measurement of a mean volatility, especially given the duration of the measurement. You have to decide if such events should be included in your measurement. Also, as sle mentioned, take into account the prevailing volatility climate for the period you are measuring (30% IV compared to 20% HV could be different from a relative standpoint with the VIX at 30, as opposed to 10). Therefore it is often quite difficult to make IV/HV bets. You can be right on with your prediction of volatility, and still lose money in many cases. There could be some value in comparing different historical volatility measurement methods for a given time-frame, and placing your data into context with volatility cones. Or when looking at options of different expiration cycles, it can be useful to measure forward volatility (or the volatility the market implies between two dates).
Thanks Doobs. I know IV30 could be up 10% while your long options' vol stays frozen. It's just a hypothetical option. I like the Sigmas better I think. Is forward vol discussed in Sinclair/Baird? Those are what I am reading next. The purpose of forward vol is if there isn't an option expiring right before an event, and you want to find the vol the market is implying for this event, then you do a calculation to come up with that vol, because just using vol a month out that has time premium distorts the calculation. Is that right?
This is a very important point that I have underestimated so far. I still study some of the models to get a better understanding of the dynamics but have less faith to their predicting ability. What you suggest is what makes the difference and the next step for me (not that my fisrst step is complete or anything lol) but seems much harder (for me)
I think it depends on how good you are at predicting randomness? That is what markets are to individuals. The markets are not random but are random to the trader that can't possibly know what price or vol is going to do, how can anyone? Successful peple either invest for term or trade size a couple times a year with catastrophic stops in so they can live another day all this day trading stuff is limited is scalability and mostly foolish Modt size option traders are working on desk where there is a degree of information that can be used bordering on inside information, deal flow. There are statistical traders that I have heard and read do well, however you won't find much here, I have tried peple here most interest in time frames and indicator bs
there are ways to make money.. people are lazy and not creative enough.. you know its like that example with random walkers speaking of a hundred dollar bill lying out where it can be found.. people expect the money to come hit them in the face..