daily returns. However, Python does not offer packages for garch(1,1 thus I think I have to implement it myself. Intuitively this makes sense, as method 1 needs at least 10 days and has only one parameter, so method 2 might need 10*3 days (4 years) for 3 parameters (assuming orthogonal etc). One thing that is missing from this discussion - only buy the straddle if your forecast for future realized volatility is higher than the implied volatility for which the straddles are currently selling.
When buying or selling your straddle, you'll want to make sure that it is delta-neutral. However, I am at loss where to start writing the garch function. This is neatly solved by garch(1,1 at the expense of two further parameters to estimate. You can then substitute the unconditional variance with the long-term sample variance of, for example, the last 4 years of daily returns. In reality, elevated volatility typical reverts to the mean rather quickly. Garch(1,1) on last year of daily returns: model with 3 parameters. It just assumes the current volatility continues.
A model for closing trading position based on, garch model with.
Model and Exit Strategy for Intraday Algorithmic Traders.
Python code implementing, garch.
I use method 1 in my trading, but I see can it gets carried away in periods of elevated volatility. As for what other instruments work better, there are broker"d "variance swaps which are linear derivatives on the future realized variance. You substitute today's prediction of variance with the short-term sample variance of, for example, the last 10 days of daily returns. There are numerous subtleties to VIX index pricing, and all the related indices. It might also be worthwhile to look into listed ETPs on various volatility indexes like the VIX, or VIX futures themselves if you are comfortable with them. I'm only estimating one parameter, but of course, there are two implicit in the model (4 years and 10 days). That is how you pick the strike, or time the purchases. Could anyone outline step-by-step the algorithm for garch(1,1) in this case? Reading a few papers, it seems clear that the garch model does better with longer lookbacks than one year. You then run a simple linear regression to determine beta. Prediction of tomorrow's variance var_4_years beta * ( var_10_days - var_4_years ). I am trying to use garch(1,1) to find the hedge ratio as described in this paper /AFE/AFE_docs/cibef0402.pdf.