what's the difference?

Is this a trick question?

Look at my post count. I am a newbie.

ok, let me explain what I want to know.If I want a swap that pays every 3 months why is a 3m curve used and when I want a swap that pays every 6months a six months curve is used and if I want a swap that pays every 12 months, still the same curve as for the 6months is used.

It's not an issue that you're a newbie, but I just want you to be a bit more specific and provide some context.So what you're asking isn't really what the difference between the two indices is, but rather why different indices are used in different circumstances. Is this correct?

Really I am looking for an answer to both. I don't think I can understand one without the other.

QuoteOriginally posted by: younggoddiewhat's the difference?A difference between 6mLibor and 3mLibor is their underlying. If one uses for example daily compounded an overnight rate for calculation yield then 6mLibor and 3mLibor would be the same at least theoretically. For swaps spreads situation looks more complex and their theoretical coincidence could be stated in a very non interesting cases. Actually, the question sounds a little bit too broadly. It would be better to write a function notations and ask why they are not equal.

Why if you have a swap that pays in dollars is a dollar curve used, but if a swap pays in sterling a gbp curve is used?It's the same question with the same answer.

QuoteOriginally posted by: TinManWhy if you have a swap that pays in dollars is a dollar curve used, but if a swap pays in sterling a gbp curve is used?It's the same question with the same answer.I do not know the answer. When two currencies are involved in valuations it might be there is not one perfect answer. If there is no arbitrage both currencies could be used for payments even if one can be logically preferred. It also confuse me even in more simple situation: FRA at t nm the payoff at T = t + n + m is ?broadly? N ( L - f ) ; N = notional, L = labor , f = forward rate defined at t. We discount this payoff to the date t by the discount factor generated by Libor rate. It does not look incorrect but I do not see convincible an argument in favor, ie argument that rejects any other possibility to discount.

QuoteOriginally posted by: TinManWhy if you have a swap that pays in dollars is a dollar curve used, but if a swap pays in sterling a gbp curve is used?It's the same question with the same answer.The problems 1. "6mLibor.USD vs 3mLibor.USD" and 2. "Why if you have a swap that pays in dollars is a dollar curve used, but if a swap pays in sterling a gbp curve is used? It's the same question with the same answer " looks somewhat different.

QuoteOriginally posted by: list The problems 1. "6mLibor.USD vs 3mLibor.USD" and 2. "Why if you have a swap that pays in dollars is a dollar curve used, but if a swap pays in sterling a gbp curve is used? It's the same question with the same answer " looks somewhat different.That's because you don't understand the question.It's very simple. Why would you imply a 1m forward rate from a 3m FRA rate?Answer, you wouldn't.

TinMan could you please to point me where you found that I " imply a 1m forward rate from a 3m FRA rate? "

QuoteOriginally posted by: younggoddieok, let me explain what I want to know.If I want a swap that pays every 3 months why is a 3m curve used and when I want a swap that pays every 6months a six months curve is used and if I want a swap that pays every 12 months, still the same curve as for the 6months is used.Hmmm Ill give your question a crack - I am sure MartinGhoul and TinMan can give you much more comprehensive answers..What you're saying is not strictly correct. If you want to price a derivative which depends on 3m LIBORS you need a 3m forwarding curve and ALSO a discounting curve depending on the credit quality etc of the counterparty at the moment I think EONIA discounting in the EUR market is the standard for counterparties with a 2way CSA. You make the EONIA curve from OISs.Likewise for a 6m dependant derivative you would still use your EONIA curve for discounting. You obviously cant (anymore) say that if I have a 3month curve then essentially I can deduce 6m forwards due to the existence of significant basis spreads.i.e. (1 + 0.25 * spot 3m LIBOR) * (1 + 0.25 * 3m forward 3m LIBOR) is NOT equal to (1 + 0.5*6m LIBOR)In the EUR market I am quite sure that to price a derivative dependant on 12m EURIB you dont use a 6s curve but this might be different for USD, I am unsure as to why you would use a 6s curve to do this - maybe the above mentioned basis spreads (6s12s) are negligible?

QuoteOriginally posted by: list TinMan could you please to point me where you found that I " imply a 1m forward rate from a 3m FRA rate? " Where did I say that you did?QuoteOriginally posted by: DocToc What you're saying is not strictly correct. If you want to price a derivative which depends on 3m LIBORS you need a 3m forwarding curve and ALSO a discounting curve depending on the credit quality etc of the counterparty at the moment I think EONIA discounting in the EUR market is the standard for counterparties with a 2way CSA. You make the EONIA curve from OISs.Likewise for a 6m dependant derivative you would still use your EONIA curve for discounting. You obviously cant (anymore) say that if I have a 3month curve then essentially I can deduce 6m forwards due to the existence of significant basis spreads.i.e. (1 + 0.25 * spot 3m LIBOR) * (1 + 0.25 * 3m forward 3m LIBOR) is NOT equal to (1 + 0.5*6m LIBOR)In the EUR market I am quite sure that to price a derivative dependant on 12m EURIB you dont use a 6s curve but this might be different for USD, I am unsure as to why you would use a 6s curve to do this - maybe the above mentioned basis spreads (6s12s) are negligible?Just read this, that's pretty much all there is to it.

QuoteOriginally posted by: TinManQuoteOriginally posted by: list TinMan could you please to point me where you found that I " imply a 1m forward rate from a 3m FRA rate? " Where did I say that you did?QuoteOriginally posted by: DocToc What you're saying is not strictly correct. If you want to price a derivative which depends on 3m LIBORS you need a 3m forwarding curve and ALSO a discounting curve depending on the credit quality etc of the counterparty at the moment I think EONIA discounting in the EUR market is the standard for counterparties with a 2way CSA. You make the EONIA curve from OISs.Likewise for a 6m dependant derivative you would still use your EONIA curve for discounting. You obviously cant (anymore) say that if I have a 3month curve then essentially I can deduce 6m forwards due to the existence of significant basis spreads.i.e. (1 + 0.25 * spot 3m LIBOR) * (1 + 0.25 * 3m forward 3m LIBOR) is NOT equal to (1 + 0.5*6m LIBOR)In the EUR market I am quite sure that to price a derivative dependant on 12m EURIB you dont use a 6s curve but this might be different for USD, I am unsure as to why you would use a 6s curve to do this - maybe the above mentioned basis spreads (6s12s) are negligible?Just read this, that's pretty much all there is to it.If reads the statement " A difference between 6mLibor and 3mLibor is their underlying" then it can be formally expressed by the formula "1 + 0.25 * spot 3m LIBOR) * (1 + 0.25 * 3m forward 3m LIBOR) is NOT equal to (1 + 0.5*6m LIBOR)" but I could not understand the sense of "USD" in XmLibor.USD.