Forums  > Trading  > Volatility Trading And Predicting  
     
Page 5 of 5Goto to page: 1, 2, 3, 4, [5] Prev
Display using:  

Strange


Total Posts: 1597
Joined: Jun 2004
 
Posted: 2019-02-16 23:45
"the name of the game these days is disentangling costs of events, not just gammas"
It's been going on for years. In SNO (and to some extent, in FX and rates), the upcoming events get priced in by MMs rather early on. The models for it are not that fancy.

"In Russia, every CDS ends in bullet payment"

frolloos


Total Posts: 86
Joined: Dec 2007
 
Posted: 2019-02-17 07:48
What is meant with "cost of events"? Is it another name for jump risk?

One man's Theta is another man's Gamma - Me

Strange


Total Posts: 1597
Joined: Jun 2004
 
Posted: 2019-02-19 04:56
Yeah, something like that. More along the lines of "expected volatility across a specific event" not necessarily a jump in the pure sense of that word.

"In Russia, every CDS ends in bullet payment"

djfostner


Total Posts: 29
Joined: Oct 2008
 
Posted: 2019-04-04 01:33
I've spent most of my time in the commodities space so I apologize if my lingo isn't in line with the latest jive, but as I've seen the concept applied over the years I'd say it's less related to "jump risk", which can often just be accounting for fat tail expectations in the distribution of the underlying, and more about expectations/value of a particular event/event type.

Considering an example of a nat gas vol book...I'd consider "jump risk" to be the uncertainty of a cold streak in winter or hurricanes in summer and at least years ago, the result was the call skew probably would seem a bit rich to some, and many industry vets made a habit of paying up to keep long units way up the call wing. "Cost of events" on the other hand meant more along the lines of theta bills for all times are not created equal. If I'm taking out time premium via "days to expiration" it might make sense to consider methods other than simply taking out 1 each day. Consider if someone were to ask you to price a straddle that expires 2 trading days from now every day for a week. Even if the market is firm and consistent in it's implied IV and the underlying plays nice so each day the market is comparable when you're asked for your quote, you should still probably have some fairly different answers during that week depending on what that 2 day horizon includes at the time of the quote. Does the underlying have a pronounced weekend effect where the extra weekend days included in the 2 trading days duration mean a Friday -> Tuesday straddle has a greater expected vol than Monday -> Wednesday would (common in weather sensitive products)? In the case of energy products, is an DOE/EIA inventory/storage number or other financial data announcement included in the period that would could possibly contain market moving information? Is OPEC meeting? Has The Donald been especially nimble fingered with his recent tweets? Basically it'd be tough to make it as a MM in many different products without being somewhat competent in being able to value the optionality of various specific event types and their varying significance throughout the year (esp so in single name equity options where earning announcements seem to drive some very interesting vol surfaces. Vol Dynamics has some available research available on some of the less common surfaces for those interested).

Unfortunately I really don't know of any good sources about the practice, but that might be due to the nature of the problem. While I've seen advances since I first saw the concept applied in pricing FX options in '08/'09, with even some mostly systematic methods/models in production in pricing various FI vol products, at least in commodities the process seems to still involve a fair amount of domain knowledge reasoning and plenty of magic hand waving.

tbretagn


Total Posts: 281
Joined: Oct 2004
 
Posted: 2019-04-04 09:50
"cost of events" is simply, if I calibrate my 1 month volatility, and on the 1 month date we now have say the Brexit referendum, I should add a premium to my usual calibration to take into account this very special event.
Tomorrow that 1 month date reverts to a more "normal" volatility (think NFP on friday, the monday is not as volatile).


Et meme si ce n'est pas vrai, il faut croire en l'histoire ancienne

Jurassic


Total Posts: 277
Joined: Mar 2018
 
Posted: 2019-04-04 22:44
why is it more important to get the events and not the bits inbetween the events?

tbretagn


Total Posts: 281
Joined: Oct 2004
 
Posted: 2019-04-05 12:49
Because moves are roughly linear in-between events, and your distribution lognormal to a certain extent.
While the event is a jump by nature.
So if you were to do a pre-post trade, you end up pricing that event volatility itself.

Et meme si ce n'est pas vrai, il faut croire en l'histoire ancienne

henderson


Total Posts: 178
Joined: Jul 2007
 
Posted: 2019-05-12 17:02
Did you every come by a copy of the paper? I'd be interested in reviewing a piece of history as well if so.

Jurassic


Total Posts: 277
Joined: Mar 2018
 
Posted: 2019-05-15 19:14
"Because moves are roughly linear in-between events"

@tbretagn markets dont move in a straight line between events?


Vishy Anand


Total Posts: 1
Joined: Jun 2019
 
Posted: 2019-06-30 09:36
Hello Friend,

Could you please send me a copy of this paper too.

Thanks,
Anand

cherk


Total Posts: 1
Joined: Aug 2019
 
Posted: 2019-10-20 00:55
Can I get a copy as well? Thanks

PicturePerfect


Total Posts: 2
Joined: Feb 2018
 
Posted: 2019-11-20 00:08
Would anyone be able to graciously share a copy of this with me as well? Thank you.
Previous Thread :: Next Thread 
Page 5 of 5Goto to page: 1, 2, 3, 4, [5] Prev