[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: MKT:Options "FAT TAILS"



PureBytes Links

Trading Reference Links

Donald Thompson wrote:
> 
> Dr OEX,
> 
> You talked a a couple of days ago. about the existance of the socalled
> "FatTails" existant on the "normal" distribution curve.  And the danger
> of running into trouble near term..
> 
> Do you think anyone has solved for these tails? In a formate similiar
> to  the normal distribution curve.  Or is there some other key words out
> here that might hint at how to approach the examination of them.
> 
>  I guess at a minimum a person should at least look at the places where
> they have occured and study the price action surrounding them..
> 
> Thanks in advance

I saw the earlier post about fat tails but I actually wasn't sure what
the point was. So I am not sure if I am answering your question, but my
two cents..
Fat tails are not new at all, they have been well known (in futures
options) back to at least 1985 that I know of.  Granted, it took some
time before option models were able to accurately model them, but that
was also some time ago.  I became an option market maker in 1989 and (at
least on the futures options side) our firm was well along in "Solving"
them.  Obviosly black scholes does not address fat tails, but there are
any number of models widely available that handle them well.  These
models are sometimes known as "pump and skew" models or even
"alpha/beta" models (which is not really correct).  These models can
create virtually any distribution curve seen in option trading with the
exception of stock options of a takeover target.
The concept of fat tails in simple terms (sorry to those who already
know this stuff) is just that markets actually make large (i.e 2 and 3
standard deviation) moves more often than normal distribution says they
should, so, way out of the money options should be worth more than BS
says they should be.  The problem that I saw over the years with
modeling these tails was not how to get the theoretical value up (that
is easily forced) but how to get them up without making the deltas
wrong.  There were times in the pit when my sheets showed deltas that
were very different than someone elses (i.e 10 vs 16, that is alot on
1000 options!) People struggled for awhile with that, but there is more
"general" agreement on deltas nowadays, so I guess what I am saying is
that fat tails have been "solved for" and and they are generally priced
fairly in any liquid marketplace.

Eric