Correlation is highly portfolio dependent, so you cannot simply enter iTraxx base correlations in a bespoke tranche pricing. At best, you can use the skew of standard index tranches to gauge the market's risk aversion w.r.t. several tranches.

<t>Erstwhile is correct in saying that the risk-neutral expected value of the CPPI strategy equals initial_value*exp(rt).However, vega is the partial derivative of the actual value of the strategy, not of the expected value of the strategy. In the CPPI case (for m <> 1) the portfolio allocation to t...

<t>QuoteOriginally posted by: AaronThis is a perverse situation. If you set a risk limit, you encourage people to maximize return on risk. That is indeed what we'd like to do; whether one uses VaR or other measures of risk or cost of capital is not so crucial here. It's similar to a portfolio proble...

<t>Working in an energy trading group we face the following problem: the company has set aside some risk capital (in the form of a VaR-limit) for the entire trading group. Now we need to distribute this total capital over the separate desks (oil, coal, gas, power). How do you do this in a rational w...

if i'm not mistaken bdt stands for Black-Derman-Toy

<r>You can find an excellent series of papers on different models for the term structure of interest rates and their implementation in Mathematica on the homepage of Zvi Wiener:<URL url="http://pluto.mscc.huji.ac.il/~mswiener/research/research.htm"><LINK_TEXT text="http://pluto.mscc.huji.ac.il/~mswi...

<t>Caminus' approach indeed uses dynamic programming with Monte Carlo simulation. As a quant structurer I prefer a tree model though. You can consider storage as a Bermudan option (you can exercise every day) with 3 instead of 2 exercise options (fill up, leave unchanged, or withdraw) and depending ...