Bonders/Swapers,is it correct to price a simple Bond off 2-separate curves for P&L attribution Purposes. @ time (T ) price the bond of curve 1@ time (T-1) price bond of curve 2underlying instruments of curve 1,2 different .is this remotely correct.Thanks,

What are the curves 1 and 2?

curve (1) based on EUR Deposit rates and swaps.curve (2) based on EUR deposit rates ,EURodollar futures,and swaps.example (2)or as curve (1) above,and curve (2) is Eonia curve.thanks in advance.

Well, that would be a completely wrong thing to do, IMHO. I can't see any possible reason why you'd want to do this.

Thanks Martinghoul .you are absoultely correct. well it is not me who want to do this ,but it is something that came up in the implementation of a system,and they wanted to add this as a risk fator when generating the P&L report,i said it is wrong,but i wanted to get a more enlightened and sound jugment from veteran in the fieldis that wrong in both cases above?the curves are 2 completely different creatures with different set of assumptions.

QuoteOriginally posted by: SamsaveelThanks Martinghoul .you are absoultely correct. well it is not me who want to do this ,but it is something that came up in the implementation of a system,and they wanted to add this as a risk fator when generating the P&L report,i said it is wrong,but i wanted to get a more enlightened and sound jugment from veteran in the fieldis that wrong in both cases above?the curves are 2 completely different creatures with different set of assumptions.Well, it doesn't make sense, because I don't really understand what it means to "price a bond off of two different curves".

what I mean is using the discount factors derived form bootstraping the above curves to get the PV of the bonds future cash flows.and then evaluating the difference in the PV's to arrive at the P&L .

QuoteOriginally posted by: Samsaveelwhat I mean is using the discount factors derived form bootstraping the above curves to get the PV of the bonds future cash flows.and then evaluating the difference in the PV's to arrive at the P&L .there is nothing wrong with this so long as the bonds are of the same credit quality. perhaps you can only do this if the bonds are default free.

so it will only be correct if the bonds are default free Ahhha ,otherwise the cash flows should be discounted from the same curve at 2 different points in time.is that right?thanks in advance

QuoteOriginally posted by: Samsaveelso it will only be correct if the bonds are default free Ahhha ,otherwise the cash flows should be discounted from the same curve at 2 different points in time.is that right?thanks in advanceThat doesn't sound right to me. The two curves are completely different in all aspects. How is this a fair P&L attribution regardless of credit quality?

Hi all, I am pretty new to Wilmott forums. The thread's gone silent, but I guess I could put in my thoughts here:You can most certainly have the two-curve framework proposed - one for pricing the other for pnL assessment. I understand this is a very standard practice, as rarely would a trader's curve be the same as that which the control guys use to calculate day to day pnl. Pricing curves could be very similar in build to the pnl curve, but can have overwrites to present a more personalized view on market factors (for example rate-hikes expectations). To put it simply, for any goods in a market there's a market price at which the market can quickly absorb it - and there is a price at which YOU can make the goods available, i.e., trade it. They need not be the same at all. However, to assess how well you do each day, you will always benchmark your performance against the market prices as described above. Usually the pnl curves are very generic in build and are a direct reflection of most liquid available market instruments. A pricing curve, however, can take into account things such as bumps to account for liquidity tie-ups on turn-days, the rate-hike expectation of the trader which may differ from the market etc.