wilmott.com

hi all,im trying to implement the exponential default intensity model propsed by Giesecke (2003), using 2 or 3 names.Part of this involves calculating the risk neutral default probabilities. Im trying to do this using the Hull&White method from their paper 1 where the difference in price between zero tresaury and zero corp is the default value and discounting etc....The problem is that I need to use zero coupon bond data and i can only find coupon bearing bond data for corps.So I need to know how to find the zero coupon rates from the coupon bonds - i know that a coupon bond can be divided up into zero coupon bonds but I don't know what rate to discount at etc...it's probably an amateur question but i've not had much experience with doing this stuff.I have treasury zero coupon bond data from Yahoo Finance but corps dont seem to issue them.I've heard about bootstrapping - but ive not seen the method and i dont know if its too time consuming for what is only part of the overall model.Also, the corp coupon bonds have widely varying maturities, which I need to match with the more regular treasury zero data - is it OK to simply piecewise interpolate beween the corp zero data once i calculate them??Any help would be appreciated Zak

If I read your message right, you want to estimate the zero coupon term structure and use that as in input to your model? If so, I would say that it wouldn't be a good idea to use corporates in your sample because theyield curve you're going to estimate should hold credit constant throughoutthe term struture, if you mix corporates with riskless at one end of the termstrusture, your yield curve is going be biased.I had the same problem when I was doing some research on yield curveestimation. If I was you, I'd skip the whole process of estimating the yieldcurve and just get the data for the term structure from another source. I'mpretty sure you can download zero coupon yield curve data from bloomberg atvarious frequencies. You can also take a look at Hu McCulloch's website as well.He has lots of data and his methods are widely known. Other than that, if you want to estimate it yourself, you should be able to access the data through CRSP at daily and monthly frequencies. Actually, I think thatCRSP has the same data as McCulloch does but don't quote me on that! CRSP will give you allthe quotes that you need, then you'll have to choose a method like spline,exponential slpine, Nelson-Siegal etc. It will add alot of work to whatever it isthat you're doing so again, I would recommend that you get the data fromanother source.r

r2338 - see pvt msg please