ok thanks, I was evaluating it numerically myself and cutting off too early ([0-l] l too small)

Did you manage to evaluate the integral numerically?I can't find the same result as Gatheral for the same parameters, my result seems to be very instable.What method do you use ?

<t>QuoteOriginally posted by: erstwhileKinda interesting: Sorin R. Strata gives an expression for the price of a european call option, with additional terms due to skew and kurtosis (based on Gram Charlier). The Gram-Charlier series uses the moments of the distribution, and the Edgeworth uses the cu...

ErstwhileI think you are talking about gram charlier or edgeworth expansion.in the formula below , s is for skewness and k for kurtosis, n is the gaussian(0,1) densityG

maybe?

No, the payoff of a capped variance swap is not the same as a call spread.

bringiton,a capped variance swap is a variance swap minus an otm call on realized variance struck at the cap level.G

<t>"Size? You can probably trade $500,000 vega"I was more asking about how wide the bid offer is (eg dec05 ATM Call would be worth around 2.0 at the moment, but how wide the bid offer ?)"But implied vol is far "jumpier" than realised vol. All it takes is some little jitter and up it pops!It seems to...

<t>Erstwhile,this is very interesting and raises a few questions:"When I first started trading these things there were only two market makers but there must be at least five now."How large are the markets they make ? (SPX dec05 or DEC06 ATM for example)"Note that options on implied variance (the opt...

<t>"You are looking at the old "correlation based on 90% strikes is 65%, but based on 110% strikes is 45%" business?If so that seems a complicated and highly path dependent way to try to make money."Yes but not only, I hope it will shed light on the dynamics of a correlation matrix as well."The late...

Very interesting, thanks,I am more interested in this problem from a dispersion / relative value of index with respect to components point of view.Don't know if this is going to lead me anywhere though.Greg

<t>Thanks erstwhile,"I managed to get both index vol surfaces reasonably fitted,The product ..."Let me see if I understand what you were doing:Your simulation was fitting market prices for each underlying separately (single stock or index options).But then you wanted to price some multi (2) underlyi...

<t>...in a stochastic vol model where there is a correlation between spot and vol (eg Heston...)Having two sets of heston parameters, one for each stock with each their own correlation between brownian of spot and vol, what correl between the brownians of the 2 stocks to use so that the measured sta...

thanks

Anyone got this?G