Volatility Trading + CDROM by Euan Sinclair, Ph.D.

filthy · Post by **filthy** » Thu Apr 18, 2013 2:12 pm

i wouldn't buy it on kindle. and i love my kindle but i've never been happy with graphs on it. the equations don't bother me as much. i also find it hard to jump from section to section which is how i normally use books like this.

but thanks for ordering it in any form.

TonyC · Post by **TonyC** » Sun Apr 21, 2013 10:38 pm

got the book thursday but front desk didnt get it to me till friday afternoon ... spent spent saturday at mollys reading it whilst sipping frozen irish coffee (google mollys in "new orleans mon amour")

the tattooed bartendrix and the assitant chief of police showed enough interest that that i ordered em some copies

HankScorpio · Post by **HankScorpio** » Wed Oct 16, 2013 2:40 pm

Ladies and Gents, my apologies if I'm being a bit thick here but I have a question regarding equation (7.15) on page 126 of the second edition of Euan's book (the one timestep mtm profit if we hedge using implied vol):

[img]/User%20Files/2641/Latex-Equation-11823.gif[/img]

all is fine, but I cannot see how the realised variance comes into the equation on the RHS. It would appear that the Gamma term from C(sigma_i) must be:

[img]/User%20Files/2641/Latex-Equation-11824.gif[/img]

but why would the Gamma (and all other Greeks) be evaluated using implied vol, whilst the multiplier for this Gamma term is the realised variance?

Many Thanks.

aschon · Post by **aschon** » Wed Oct 16, 2013 4:26 pm

If you replace dC by

[img]/User%20Files/4661/Latex-Equation-11825.gif[/img]

the realised variance comes into play, since

[img]/User%20Files/4661/Latex-Equation-11828.gif[/img]

Hope this helps.

[edit] bug fixed

HankScorpio · Post by **HankScorpio** » Wed Oct 16, 2013 11:08 pm

Hi aschon,

Thank you. I'm assuming that is meant to be an S^2.

However, this still doesn't look right to me, because it then loses the dt term. Confused

[edit] actually it makes perfect sense. The dt term has to exist due to scaling the variance from the (dS/S)^2 term. Thanks aschon, that helps heaps.[/edit] Beer

aschon · Post by **aschon** » Thu Oct 17, 2013 8:33 am

Glad to help out.

Yes, should be an S^2. Just fixed it.

swyx · Post by **swyx** » Tue Nov 26, 2013 5:25 am

hi folks - pardon if this has already been discussed (tho it didnt come up when I skimmed thru the thread) - whats the difference between Vol Trading and Option Trading (i mean the books by euan of course)? on a superficial TOC level they look similar. which is the more "advanced"?

minor disagreement about vol level dynamics leading up to an event - euan observes how it usually gets bid up to the event and this is a possible source of profit if it exceeds decay. from experience this has nothing to do with market inefficiency and everything to do with "heterogenous time" shoehorned into calendar time. i.e. if i have an event on thursday where 3x regular business day variance is priced in, then the implied vol of a friday option will be perceived to rise from monday to wednesday because I am distributing an unevenly decreasing total expected variance over a smoothly decreasing calendar time. while it is certainly possible that the market could belatedly be rushing for options in the leadup to the event I find the above explanation usually holds in practice and also explains why the "vega rebate" will generally not compensate for the decay all else equal.

incidentally - because i work in fx vol - reading Vol Trading made me realize how much we in fx vol take liquidity for granted - we basically dont really worry about strike or tenor contract liquidity issues

filthy · Post by **filthy** » Wed Nov 27, 2013 3:38 pm

"option trading" is the more basic and self contained. it is meant to be a natenberg replacement. there is some overlap but vol trading is more advanced.

wrt to earnings and vol dynamics you are missing the point. sure you can make iv

go up less by rescaling time but the central issue is that the options are mispriced

relative to the realized vol in any case.

buying options 2 weeks before earnings makes money. selling options immediately before

makes money. that is true no matter what you do to iv.

NeroTulip · Post by **NeroTulip** » Wed Dec 04, 2013 5:15 pm

@Filthy: How consistent is that across time and markets? Or is it just a "well chosen example" ?

filthy · Post by **filthy** » Wed Dec 04, 2013 8:39 pm

i think it is a pervasive effect.

it has been shown to hold in equity markets for the last 20 years but i think

the idea of buying a few weeks before an event and selling immediately before the

even is valid over most products. there was a study that showed the effect in t bonds

before NFP numbers and it also works for commodities and crop reports etc.