aliceww2012 发表于 2013-5-23 11:37 
谢谢回帖,但是您说的是经验违约概率计算的方法,我说的是在用MERTON模型的时候违约边界的计算。因为我感觉 ...
I think there is a model under the structured approach talking about modeling credit risk when there is a convertible bond, I can't remember the paper and the author's name. But I think it is just a little modification of the Merton's model.
The basic idea is that, when the equity's price goes up to a certain extent, the debt is converted into equity. So there is a boundary (K1) of whether converting the debt to equity. You can easily determine this boundary by calculating the dilution effect of the equity if the convertible bond is exercised.
In this case, the question has become something like, given a firms value VT, there is a corresponding Equity and Debt: ET and DT. It is a piecewise function. You just need to find this payoff and decomposed them into several options.( Actually, in this case ET and DT are all options on VT. In traditional Merton or KMV, there is only one option, but here you have more). Then you can price the debt and determine the default probability easily.
Just the basic idea, you should checking the model.
best,