MSc Thesis Synthetic CDOs Pricing

danongohou · September 4th, 2007, 7:55 pm

Hello ! I am not a quant nor a developper. I just feel I gained some valuable knowledge Credit Derivqtives Pricing methids thanks to Andersen lectures notes, The Merrill Lynchderivatives handbooks and Galiani (2003). I would like to share this experience.here is My thesis : Pricing and Risk Analysis of Correlation Products- Evidence ofSynthetic CDO Swaps. I attached an Excel file where I implemente a Homogenous casein light of this thesis.Ps: any comment will be appreciate since I am waiting for my result!!!!RegardsDANON

Lou2007 · October 3rd, 2007, 11:09 am

Hi DanonI tried to play with your excel model and it seems that the spreads for the low tranches (Equity, Junior, Mezz1 and 2) get higher when the correlation increases, which should not be the case. Any idea why?You must have worked hard on this thesis so I hope your results will be goodThanks

trancheitup · October 8th, 2007, 3:14 pm

I haven't looked this over in detail, but I would agree with Lou2007 at least in terms of synthetic CDOs. ALL ELSE EQUAL, as correlation increases, equity spreads should decrease (the probability of 0 losses goes up) and the senior tranches spreads should widen (the probability of large losses goes up). That is consistent with Gibson, but not consistent with your Annex 4, unless you're somehow quoting price instead of spread. In which case I'm not quite sure how to read the table. You definitely covered a lot of material, thouh! Good luck on your results!

danongohou · October 8th, 2007, 8:54 pm

Thanks a lot Lou and Trancheitup for you notes on the spreadsheet.

gbelford · June 1st, 2011, 3:57 am

ALL ELSE EQUAL, as correlation increases, equity spreads should decrease (the probability of 0 losses goes up) and the senior tranches spreads should widen (the probability of large losses goes up). That is consistent with Gibson, but not consistent with your Annex 4, unless you're somehow quoting price instead of spread. In which case I'm not quite sure how to read the table. Just wondering if anyone ever figured out an answer to Tranchitup's q. above?Thx and Regards,Grant

loooooo · July 30th, 2011, 7:50 am

Just intuitively, an increase in correlation means that one default can have an impact of a somewhat larger scale, so the probability distribution of losses now exhibits a fatter tail towards the right (i.e., the probability of large losses goes up), whereas the probability of 0 losses decreases. Hence the equity tranche becomes cheaper in the face of an increase in the spread of senior tranche.

gbelford · August 1st, 2011, 1:18 am

Thx - agree with you! But the results from the spreadsheet appear to show the opposite occurring - just trying to determine why this is?

binbintriangel · November 24th, 2011, 12:37 pm

has anyone found out the inconsistent in the excel?for UnconditionalDefaultProbability, after T1, T2 and T3, in T4, the formular is suddenly changed, directly result in smaller ProtofolioLossDistagain, in PremiumLeg, in thesis, it specify (EL0+EL1)/2 however in the excel, it reads EL0+EL1/2has anyone noticed this?