wilmott.com

Hi,I'm new on this forum and hope to get an answer to a simple question. I am trying to explain the pl of a swap portfolio is several components. I now have Delta, new trades, amendments, unwinds. The last component Theta I don't quite understand; is this imply a $number or do I need to multiply this with the yield change just like Delta to get the pl explained? Also if you guys have additions to the pl components I use please feel free to comment.

you need to multiply theta with (t2-t1) when explaining PL from t1 to t2. Also be aware that theta can be computed in several ways and that it should be consistent with your delta contribution. E.g. theta computed with constant market quotes and delta computed as market rate delta vector times market rate changes. Or theta on forward curve and delta multiplied with difference of t+1 curve from t+1-forward curve as seen from t. You can also use a gamma matrix to improve explanation.

what about convexity...? If you have a large portfolio with long dated swaps you should definitely include this to make your pnl explanation better. edit: apologies didn't see pcaspers last line!

what about resets..? if you have a big swap book this should make a significant portion of pnl daily (unless this is included in amendments etc..?)

Sorry for the broken up answer but also depending on the swap portfolio you have you will also have lots of bond futures and possibly swaps on different indices etc..So you need to also break your pnl/risk into spread risk/pnls

QuoteOriginally posted by: DocTocwhat about convexity...? If you have a large portfolio with long dated swaps you should definitely include this to make your pnl explanation better. edit: apologies didn't see pcaspers last line!how do you defind p&L due to reset or fixings ?also if you have a swap book based on 2 CCy,one being a spread to the other,how do you include the basis risk as part of the P&L ?

Say you have 1 swap in your book say it is on 10y cms and say you are paying..pnl due to fixing is = - notional *(ISDAFIX for 10y swap - 10y swap at the end of the day)*accrual for your second question i dont know much about xccy basis but just for basis you need to also have a basis delta or basis explanation likewise for pnl.another thing to keep in mind with swap books is that say you have a 10y irs this might have futures against it. so say you are rec in a 10y irs then you might have sold bund futures against this. obviously this gives you exposure to the the spread between the two rates. Now because they not exactly the same you also need to have some sort of spread pnl explanation..because to see the risk for the 10y swap you would use a euribor swap curve for the bund future you would bucket this risk into the euribor swap curve however they dont move in exactly the same way causing a spread pnl too.

If you have stuff like libor-in-arrears, changes in volatility would affect p&l.IIRC, we used to compute theta by assuming that today's zero curve was the same as the next day's zero curve and then revalued all the trades. This had the effect of moving all cash flows one day closer.

QuoteOriginally posted by: DocTocSay you have 1 swap in your book say it is on 10y cms and say you are paying..pnl due to fixing is = - notional *(ISDAFIX for 10y swap - 10y swap at the end of the day)*accrual for your second question i dont know much about xccy basis but just for basis you need to also have a basis delta or basis explanation likewise for pnl.another thing to keep in mind with swap books is that say you have a 10y irs this might have futures against it. so say you are rec in a 10y irs then you might have sold bund futures against this. obviously this gives you exposure to the the spread between the two rates. Now because they not exactly the same you also need to have some sort of spread pnl explanation..because to see the risk for the 10y swap you would use a euribor swap curve for the bund future you would bucket this risk into the euribor swap curve however they dont move in exactly the same way causing a spread pnl too.-what is the ISDAFIX.-How do you compute the accrual ?

Vanilla swaps fixings have absolutely zero to do with ISDAFIX.Samsaveel, you have too many basic questions. It's impossible and counterproductive for people here to try answering them all, IMHO. Have you read the required material, such as Flavell and Tuckman? I believe you'll be able to get a much better understanding that way.