kouhaku 发表于 2014-2-14 21:45 
谢谢回复。
昨天老师探谈过这个问题,他主张用interest rate swap 去hedge。不赞同把credit risk单独分出 ...
Ok, fair enough. IRS can do this.
But I am more concerned about the risk if the default really happens. This is the risk that the normal interest rate derivative cannot hedge, and one can't say this is impossible when he manages his Cop bond portfolio. That's why I separate the credit risk. (Roughly, treasury bond(risk free)+ CDS( Cop)=Cop bond. This is formula I am thinking about)
Of course, if you ignore this risk, your advisor is correct. The best hedging instrument is of course it's derivative. So, I am just thinking about a step further. You can ignore it :-) I am used to thinking questions in a theoretical way, so maybe mine is nonsense. The best why to find out the idea is read the existing papers on this topic.
Duration has some drawbacks, but if you take it as a local sensitivity of the bonds price respects to a small change in yield, it is a good hedge ratio. Or you can calculate the empirical hedging ratio. Since treasury and the risk free part of the Cop bond has the same risk source, it is a good hedging instrument. Many people still use BS greeks for hedging option, see how many assumptions in the BS model. :-) And you actually don't need to think duration as duration. It is just a sensitivity (I think even you use IRS, you may still need the sensitivity for hedging), Macaulay is a little useless, but modified or dollar duation is good, and are still useful for hedging.
My interest is in asset pricing, mainly on state price density (pricing kernel), but I also did many other researches in finance and financial engineering. So my interest is relatively narrow, but my research area is quite wide (most of them done with others). I am B in Econ, M in QF, and is applying for PhD in finance. Just for your reference.
Sorry to be a little wordy.
best,