Serving the Quantitative Finance Community

 
User avatar
snapperjoe
Topic Author
Posts: 0
Joined: September 8th, 2011, 2:46 am

Pricing Options with extended trading hours

September 8th, 2011, 3:39 pm

I am interested in some how quantifying the advantage of being able to scalp gamma in a product which trades 23 hours a day compared to one that trades 6.5 hours a day. An example would be two products that are nearly 100% correlated such as WTI crude futures which trade on Nymex and the ETF USO which trades on nyse. These two products are very highly correlated, however the futures trade 23+ hours a day compared to the ETF trading roughly 6.5 hours a day. It would seem logical to me that the WTI futures would offer a greater opportunity to cover theta via gamma scalping than the USO stock. But I would like to somehow quantify this to evaluate weather this could be a strategy employed over several different products, with similar disparities in trading hours. I have the capability of looking at minute bars in futures and the corresponding ETF's. I would think quite simplistically I could look at the true range for the entire session and would likely see the future with a larger range, but this will not quantify many intra-day scalps which would occur. I have put this into a back-test and can quantify gross dollar day trading profit from scalping a set percentage intervals but was wondering if there is some type of way to evaluate this near continuous trading volatility?
 
User avatar
kcross
Posts: 0
Joined: March 31st, 2011, 8:11 pm

Pricing Options with extended trading hours

September 11th, 2011, 2:30 pm

I haven't (although want to) done analysis like this - but I've found that there are some better opportunities to trade gamma in extended hours trading.
 
User avatar
exCBOE
Posts: 0
Joined: May 19th, 2010, 9:48 pm

Pricing Options with extended trading hours

September 14th, 2011, 5:42 am

Sounds close to a free lunch. A proper model of the 6 hour-trading option must take into account all known information about the underlying. That would include any information obtainable by observations made when trading is closed--- including trading in a correlated instrument. As usual, if you find that the marketplace is NOT taking that into account properly, you may be able to benefit. The first thought that comes to mind is that options on the two highly correlated products should trade at similar implied vols. If not, buy the low implied vol ones and sell the high implied vol ones. Of course the devil will be in the details of just how well they correlate and the exact terms of the instruments you trade. Good luck!
 
User avatar
snapperjoe
Topic Author
Posts: 0
Joined: September 8th, 2011, 2:46 am

Pricing Options with extended trading hours

September 20th, 2011, 7:24 pm

Ok, so I think I am wrapping my head around this problem a little more now that I have started calculating historical volatilities and believe for this particular trade the main issue is:Gamma Scalping to cover ThetaI am researching a trading style that will allow for near 24 hours trading and very low transaction costs, due to exchange membership and rebates. The strategy would be long front month ATM straddles. I am interested in a method to verify and measure that the profit made from scalping a long gamma position will cover time decay on a daily basis. Is there a formula to plug the gamma and theta values in to determine the movment necessary? I have access to 1 minute bars. To me comparing Implied volatility of the options to the realized volatility of the future may not be the answer because of the 1 minute time periods, and the nature of the trade essentially is a strategy which is based on the assumption that there will be enough overnight noise or range which will enable agressive market making / gamma scalping.
 
User avatar
bearish
Posts: 5186
Joined: February 3rd, 2011, 2:19 pm

Pricing Options with extended trading hours

September 21st, 2011, 10:48 am

I think you are suggesting that you will somehow make a positive profit on this "gamma scalping" activity of yours. That may indeed be the case, but it pretty much requires that the market price is mean reverting on an intra-day basis. This is not technically inconsistent with the assumptions of standard option pricing models, but it would require a pretty amazing risk premium process. This is probably not what you assume, so I think this comes down to a deeper misunderstanding of the equality of gamma and theta in the Black-Scholes model (up to sign, factors of one half, carry, etc.). This equality makes a position in an option a "fair bet", and this is true whether or not you carry out a delta hedging program. In particular, it is not like you are somehow systematically losing money on the option theta that you have to earn back through delta hedging (or gamma scalping in your parlance). If this were the case, why wouldn't you just forget about the option market altogether and harvest the gamma scalping profits in the underlying market on imaginary option positions all day (and night) long? In the standard theoretical models, hedging is ex ante a zero NPV activity. In the real world, hedging (in and of itself) is always ex ante a negative NPV activity, due to a variety of transaction costs incurred. Hedging is something we do because we want to limit the risk/required capital incurred when putting on derivatives positions that we think are favorably priced, ideally locking in an arbitrage profit. To the extent that "hedging" activities systematically make a profit they incorporate an element of proprietary trading that could probably be separated from the execution of hedges based on option pricing models.