<r>Looks like the solution is here:<URL url="http://www.qgroup.org.au/SFMW/Braziliian_Options_03.pdfBlack%27s">www.qgroup.org.au/SFMW/Braziliian_Options_03.pdfBlack's</URL> formula is applied to transformed forwards and strikes, rather than the original forward f and strike k.f* = (1+f)^((T1-T)/252)...

<t>Hi, I recently saw a Brazil swaption with an OIS swap as underlying.The term sheet specified the following:Let d=number of business days in the life of the underlying swap.Then the swap value at its termination is {Fixed Side}-{Floating Side} whereFixed rate Side = Notional*(1+fixedrate)^(d/252)F...

<t>If you enter a 10yr swap today, the 3-mth roll-down is (9.75yr swap) - (10yr swap)Because in 3 months' time you will have a 9.75yr swap that you had entered at the current 10yr rate.If you enter a 3-mth fwd-starting 10yr swap today, the 3-mth roll-down is (10yr swap)-(3m10yr swap)Because in 3 mon...

<t>it is American, the option pays a fixed amount when the barrier is first hit.This paper derives the one-touch price under jump-diffusion: Kou and Wang (2003) "First Passage Times of a Jump Diffusion Process," Advances in Applied Probability.They assume a brownian motion where the Poisson jump siz...

Hi,does anyone know of closed-form pricing solutions for one-touch digital barrier options, where volatility is stochastic?Thanks very much,Crisky

QuoteOriginally posted by: SinglestrandcriskyQuote instead I imposed constraints with transformation functionsDo you mean penalty functions?not sure what they are called, e.g. to find the y that maximizes y*(y-1), y>0, i would optimize for x, where y=exp(-x).

<t>using Matlab's Optimization Toolbox I managed to optimize across 65 parameters- it was to estimate a multi-factor Vasicek model with the maximum likelihood-based Kalman Filter method (see Babbs & Nowman '99 for this kind of approach)The available constrained optimization function fmincon() fr...

Hi berridos,bloomberg has historical data for at-the-money swaption volatilities of various expiries and swap tenorsTyping 'USSV1010 Index' will give you the 10y10y a-t-m swaptton vol, quoted as percent of yield (the black's or 'lognormal' vol).

<t>Hi bingfei,I came across a similar problem while doing optimizations- the covariance matrix had to remain positive definite but my optimization did not guarantee it.So I expressed the covariance matrix as the product of two triangular (upper and lower) matrices, using the Cholesky decomposition.T...

<t>Hi Dave,the 1-5yr option onto 1-10yr swap is a liquid part of the surface, particularly for the a-t-m's, so i would not hesitate to use the data you have for calibration. You might want to restrict the points to the 1,2,3,5yr expiries and 1,2,5,10yr tenors, as these are the most commonly traded (...

Associate is the next step above analyst. It's just a level to say that you've done a few years and know more than the graduate analyst in the same division. You get a payrise but the work doesn't usually change that much.

<t>With ExcelLink you can call matlab functions directly using matlabfcn() from the spreadsheet.Matlabfcn() takes as input arguments the name of the matlab function as string, followed by the input arguments of the matlab function.Works quite well, except if you have lots of matlabfcn()'s in the spr...

<t>that's a good point- i just checked using a randomized input but there was no change in speed, so Excel was calculating 1000x. (i initially noticed the speed difference between the excel and matlab-based functions when calculating a sheet full of bonds, where there were a couple hundred calcs). <...

oops, hit the reply button too many times..