Well. Honestly, I dont really know then. Have you tried different local vol surface parametrization? Maybe adding an extra term? (First fit a,b,c, then add a new parameter d, keep a,b,c fixed and fit d ... or something like that.)Let me know how it turns out.JK

<t>Local volatility models have some weird properties that are usually not really realistic in most of the markets. I'm guessing you work with some FX processes(?) here.Even though you fitted the current implied vol surface with your local vol model, the real world evolution of the vol surface as th...

That was my impression. Maybe somehow their legal treatment is set like that in the ISDA agreement?!

I dont think they would typically be treated that much different in case of a default. Would you have any specific example?

<t>QuoteOriginally posted by: katastrofaQuoteOriginally posted by: JoK78Besides, if you use the same recovery rate assumption in bootsrapping from the CDS quotes as in your CVA/DVA calculation, the final CVA number should be less dependent on your choice of R.Are you sure? What you're saying is true...

<t>Agreed. If no better data, I'd use R=40%. Besides, if you use the same recovery rate assumption in bootsrapping from the CDS quotes as in your CVA/DVA calculation, the final CVA number should be less dependent on your choice of R.Does it make sense to you? What accuracy do you need? What is the p...

<t>Thanks for the advices. They are very useful. Plugging in the sqrtProcess into PathGenerator is a good starting point.Though, I don't understand why the sqrt process is a good model for credit spreads. I saw it in Brigo, though. Probably because it would represent the observation that credit spre...

<r>Hi, you can simply do a hazard rate curve bootstrapping on your cds quotes. As a worked out example, check <URL url="http://quantlib.sourcearchive.com/documentation/1.1-1/CDS_8cpp_source.html"><LINK_TEXT text="http://quantlib.sourcearchive.com/docum ... ource.html">http://quantlib.sourcearchive.c...

Hi Fellow Quants,Does anyone know if there is any Monte Carlo based credit spread simulation in Quantlib, easy to use and merge with some other MC simulations? Maybe some examples as well? I'd like to set up some homemade CVA simulation for experimentationThanks,JK

Thanks acastaldo! This is what I thought as well. They could specify the main drivers though.JK

<r>Hi there,Would anyone know what is the mechanism for the exchange to determine the pricing range in more detail?The description on the website is very vague: "Price scan ranges ? in effect, the maximum price movement reasonably likely to occur, for each instrument or, for options, their underlyin...

<t>You can think of it in the following way:If the upfront payment didnt account for the credit riskiness of your (i.e. the option seller's) payoff at the option maturity, T, that means than your counterparty lent you money today in cash at the riskless rate that you will pay back at T by the payoff...