there is a very serious flaw in historical vols calculations and any analisys (StDev , Longnormal distro) based on HV can be very misleading, IMO. Look at the 30d HV graph from Oct to middle Nov, does anyone want to use those numbers for their future analisys ? Magically , HV goes from 100 to 30 overnight ( because the large one day % move of 34% dropped out of 30d range/calculations). BTW , "smoothing" or MA (even exponential) will not improve this picture by much. http://www.ivolatility.com/options.j?ticker=LXK:NYSE&R=1&period=6&chart=02&vct=4
taleb: sell me some s&p futures at market broker: how many? taleb: uuuh, a social amount! 'nuff said. quoted conversation while he was at bankers trust.
Talib's claim to fame was that he actually had a bunch of these options during the '87 crash, and he continues to go for a repeat. He talks about it in his current interview. These far OTM options are not available in most markets such as stocks, etc. He had them on Eurodollars -- one of the most liquid markets in the world.
I have to agree the vols on otm stock options are so high it will take another 9/11 for them to end profitably
Sigma comes from the lingo of the physicists. In terms of BS parameters it is: ln(S/X) sigma = ----------- s sqrt(t) where s is volatility, S is price and X strike as usual. In other words, is it yet another measure of moneyness. When Taleb speaks of 10 sigma, he means that BS delta = N(-10) ~10^-44 Since delta is also the probability of expiring in the money, the mean time to see this option expire in the money is much longer than the age of the universe. In the real markets, such things do happen, and payoffs from such bets can be very large but where Taleb is wrong is that he claims the one-time payoff will be greater than all the other times when you lose. There is sound evidence, the buyer of naked puts comes out losing even after including the crash of 1987, depression of 1990, mexican crisis of 1995, asian crisis of 1998, bear market of 2002, and Sept 11. Remember , all these happened in only 20 years. Taleb is warning us: things might be much worse in next 20 years... his lastname by the way, is shared with a certain pessimistic organization, is that a 10 sigma event too? PS: By the way, a good approximation to the BS formula for a call is C_BS = (S-X) N(sigma) + s * phi(sigma) where phi is the gaussian : phi(x) = 1/sqrt(2 Pi) exp( -x^2/2) and N cumulative gaussian N(z) = int_inf^z phi(z) dz
Dear Mr. Taleb: How does it feel to be the flavor of the month? Beware the other side of the bell curve.
Taleb told Trader Monthly interviewer that he bought options over ten sigmas out. Anyone who has traded options knows this is bullshit: super deep-out-of-the-money options have no market and, hence, prohibitively wide bid/offer spreads; even liquid treasuries would have an over 200% mark up on those kinds of options (i.e. 2 ticks by 7 ticks). That option market makers might not be able to accurately price in extraordinary events is banal; evidently Taleb doesn't know that MMs have priced OTM options with "fat tails" ever since options traded; but, one shouldn't expect an ivory tower academic who can't trade to know this. By the way, Taleb never made a dime trading, and his hedge fund vehicle, Empirica LLC, has been liquidated.
Sigma ... Yes well an approximate analogy ..... I'm not sure who trades options this far OTM .... unless you have a gigantic position to hedge ... and even then it almost makes no sense to me ..... One thing I have learned is that if I cant make sense of someones argument (after a reasonable attempt at giving it the benefit of the doubt through several alterative thesis) then I just ignore it .....
In the rates markets is it not uncommon to see caps struck at 10%, usually as CMS hedges, but sometimes outright. I have also seen (and done) risk reversals rather far out (5 delta), as they have some nice rolldown properties. There's also a lot of people out there doing a variety of disaster-scenario trades, i.e. CAT bond buyers etc.