<t>It has been a while since I read the H&W paper you mentioned but I thought that the models are similar in spirit, at least for Gaussian random variables. H&W circumvent the need for numerical integration with the evaluation scheme they propose but the results should be the same, again und...

Bloomberg or websites of I-Banks.Rgds

<t>Cross-subordination makes a CDO^2 or CDO of ABS effectively behave like a vanilla CDO with one exception. Once the subordination in the inner CDO is completly wiped out losses will occur much faster than in a vanilla CDO because of the overlaps of single names in the inner CDO. One default can hi...

<r>For recurison methods the update is a bit more involved. You have to adjust the calculation of the conditional PDs and to integrate over one additional dimension because of the additional Chi^2 variable introduced via the Student-t copula. Andersen, Sidenius and Basu wrote all that down in "All y...

If you need more information about the change of numeraire check Brigo & Mercurio's "Interest Rate Models". It contains a lot of information about this subject.

What exactly do you mean by "implied correlation" ?

<t>QuoteOne explanation goes as follows : "time decay : tranches become more senior as the expected portfolio losses decline because of shortening maturity"This is correct since the probability of default decreases over time.QuoteThe other says : "theta : the passage of time tends to increase the va...

N,let me know when you release some trade secrets and publish your first book. Should be in a line with Andy Kaufman.

N,nope, I wasn't clever enough for the army so I ended up at a bank. Thanks !

N, what are cyclotomic fields ???

<t>Just one more point. A Banach space is endowed with a norm which in turn induces a metric. Convergence of cauchy sequences is measured with respect to this metric with means that the Banach space is complete with respect to a certain norm. Things might be completly different if you change that no...

<t>The corp bond yield spread is also affected by other factors like liquidity (or it's absence). I agree with quadrature, for most liquid bonds the CDS price and the corp bond yield spread should be approximately the same. For less liquid bonds I would expect a bigger difference.You can proxy the v...

<t>QuoteOriginally posted by: noufaVery clear DDoom.It is an intersting approach.And it works. One can show some more interesting properties of the solution of the SDE but the one I told you is the most important one. It's more probabilistic stuff, I dunno how much background you have and how intere...

<t>Fermion, I haven't try to solve this problem but form my experience I would say that simmulated annealing should lead to the same result. 500 to 1000 iterations should be enough to find a proper starting point.Can your genetic algorithm handle problems of high dimension (>100, say) ? This would b...

<t>Nofua, I have some material but it is in German. I try to give you a sketch of the algorithm.Suppose you want to minimize f and assume that f is smooth. Choose some \epsilon > 0. Then solve the following SDEdX_s = -\grad f(X_s)ds + \epsilon dW_swith \grad denoting the gradient of f in x and W_s s...