Hi a quick question on a practical issue:suppose 1w eonia swap (EUR) is 1.28%. Today is 20 Jun 11. I would like to calculate df implied in this market quote with respect settlement date.So,:today: 20 Jun 11settlement date: 22 Jun 11 (t+2)end date: 29 Jun 11paydate 30 Jun 11 (end date +1)DF = 1 / [ 1 + 1.28% * (7/360)] = 0.9997511....Q1: Is DF found referred to payment date (30 Jun 11) or to end date (29 Jun 11) ?Q2: if DF (in Q1) refers to payment date, how to price 1Wx2W from 1W and 2W quotation? To price 1Wx2W I need DF on start date of my 1Wx2W (i.e end date of 1W ois - 29 Jun 11), but from 1W OIS I get the DF of its payment date (30 Jun 11)... not so big difference in practice .. let's say negligible...but to get the point..Thanks for your time

Hi,I'm no expert, but since there have been so many views of your post without reply, I will offer the obvious:IMO, interest is accrued over 7 days, so the DF applies to the end date. Isn't payment date irrelevant for the interest accrual and so the 1Wx2W pricing should ignore the payment date?

hi,the definition of the instrument tells what the relevant period is. a loan with initial payment on d1 and final payment on d2 gives you df between d1 and d2.a fra or ed future gives the estimate ( with adjustment sometimes ) for a futures libor fixing, which is for another period of d1 to d2. this is not tied to payment date for that instrument ( nor fixing date ), because if the trade is at the market, the expected payment is 0, and the market price does not change with the payment date.back to the ois swap, it is similar to the second example. just look at the definition of ois index fixings involved and you know the start/end date.

thanks a lot the definition is very simple start t+2end t+2+1wpay t+2+1w+1My question is: does the market quote allow you to calculate DF on end date or pay date?It is a bit differnet in boostrapping process in the very short end so:suppose 1w eonia swap (EUR) is 1.28%. Today is 20 Jun 11. I would like to calculate df implied in this market quote with respect settlement date.So,:today: 20 Jun 11settlement date: 22 Jun 11 (t+2)end date: 29 Jun 11paydate 30 Jun 11 (end date +1)DF = 1 / [ 1 + 1.28% * (7/360)] = 0.9997511....Question:In my DF curve:DF (29Jun11) = 0.9997511.... orDF (30Jun11) = 0.9997511

The floating leg detail is what really matters here. the fixed leg is just the market price for the floating leg.I think for 1w eonia there are only 5 fixings, so either the 22jun fixing or the 29th fixing is not included. Therefore I am a bit surprised by your start and end dates. Let's assume that the 29th fixing is not included. Then the floating leg for your example should be the start date is 22jun, which you will take the 1st o/n fixing, the fixing rate is for the 1d period from 22jun to 23jun. etc etc. until the last fixing on 28jun for the period between 28jun to 29jun, so the df following your computation would be a discount factor from 29jun to 22jun, not to 20jun.

QuoteOriginally posted by: XQThe floating leg detail is what really matters here. the fixed leg is just the market price for the floating leg.I think for 1w eonia there are only 5 fixings, so either the 22jun fixing or the 29th fixing is not included. Therefore I am a bit surprised by your start and end dates. Let's assume that the 29th fixing is not included. Then the floating leg for your example should be the start date is 22jun, which you will take the 1st o/n fixing, the fixing rate is for the 1d period from 22jun to 23jun. etc etc. until the last fixing on 28jun for the period between 28jun to 29jun, so the df following your computation would be a discount factor from 29jun to 22jun, not to 20jun.why are you surprised by may start and end dates? I applied standar market convenctio for EUR OIS.I never said the discount is to 20 Jun. To be more precise I said today is 20jun but settlement is 22jun. I discount to 22jun (to use only info coming from the quote... obuisly knowing O/N T S it easy getting DF to 20jun.. but it is not my question)Sorry probably not clear in my question. May question is on the end discount.DF (from 22jun11 to 29Jun11) = 0.9997511.... orDF (from 22jun11 to 30Jun11) = 0.9997511If I understood your answer you are suggesting that the right answer is:DF (from 22jun11 to 29Jun11) = 0.9997511 is it?

The start/end dates are indeed market convention for interest calculation period for the swap. I was thinking about the fixings start/end date, which does not include the last date. sorry about that.My point is just that the relevant period for the df is determined by the underlying fixing rates's definition. If you compute a discount factor this way for 1w swap on 10y cms rate, then the discount factor is more related to the 10y period and not the 1w period. In your example, yes, it is DF (from 22jun11 to 29Jun11) = 0.9997511. but it is to 29Jun not because the interest calculation end date for the swap is 29Jun, but because you have 5 connected 1 open day fixings and collectively they form a continuous period from 22jun to 29jun.

QuoteOriginally posted by: XQThe start/end dates are indeed market convention for interest calculation period for the swap. I was thinking about the fixings start/end date, which does not include the last date. sorry about that.My point is just that the relevant period for the df is determined by the underlying fixing rates's definition. If you compute a discount factor this way for 1w swap on 10y cms rate, then the discount factor is more related to the 10y period and not the 1w period. In your example, yes, it is DF (from 22jun11 to 29Jun11) = 0.9997511. but it is to 29Jun not because the interest calculation end date for the swap is 29Jun, but because you have 5 connected 1 open day fixings and collectively they form a continuous period from 22jun to 29jun.Thanks a lot for your time. Ok it is clear you answer. The raising question than is: will pay date lag have any impact in pricing? (I think yes..)In the above example if the ois will pay with 0d lag or 1d or 5d lag after end date apparently you will always consider df coming from quote as DF (from 22jun11 to 29Jun11).

Pay dates have to be considered. You will discount your cashflow from you paydate not end date as your aim is to find what that cashflow is worth today ... and the cashflow hits with a paydelay.

QuoteOriginally posted by: CRMsquaredPay dates have to be considered. You will discount your cashflow from you paydate not end date as your aim is to find what that cashflow is worth today ... and the cashflow hits with a paydelay.I agree with you. This is why I have Q2:Q2: if DF (in Q1) refers to payment date, how to price 1Wx2W from 1W and 2W quotation? To price 1Wx2W I need DF on start date of my 1Wx2W (i.e end date of 1W ois - 29 Jun 11), but from 1W OIS I get the DF of its payment date (30 Jun 11)... not so big difference in practice .. let's say negligible...but to get the point..