Serving the Quantitative Finance Community

 
User avatar
dimitri
Topic Author
Posts: 0
Joined: July 14th, 2002, 3:00 am

Day count basis

June 14th, 2006, 5:08 am

I have some problems understanding the difference between Day Count Basis in the Money market and the Bond Market. If we take a Swap I know that in the Money Market the floating leg of a swap is using a Act/360, which mean I calculate the year fraction of a period and the year fraction * Fix-ing-rate is my payment at a specified term date. If I changed the Day Count Basis to Act/Act, I would calculate the year fraction as the “number of days in the term period” divided by “the number of days in a year”. However if I am using the Bond market and I’m still using Day Count Basis Act/Act, my interest payment would simple be the coupon divided by the number of terms. My question now is do we calculate the Act/Act on a term basis in the Bond Market, say “number of days in the period” divided by “number of days in the period” divided by “number of terms”. As far as I know this is the method used when we are calculating accrued interest rate.I have desperately looked for book, where I could read about these problems but I haven’t been able to find one. Could someone please help me answering the problem or a reference? The reason why I’m asking is because I need to compute an asset swap using Bond market conventions.
 
User avatar
DavidJN
Posts: 269
Joined: July 14th, 2002, 3:00 am

Day count basis

June 14th, 2006, 11:37 am

As you noted, vanilla bonds pay equal coupons regardless pf the day count used. What the Act/Act daycount does is allow one to pro-rate the equal coupon for settlement between coupon periods. The accrued for any settlement date between coupon dates per $100 par is100*(coupon rate)/(payment frequency)*(days between settlement and previous coupon date)/(days in current coupon period)The definitive resource on this topic is the SIA Manual (Volume 1) which, incidentally, has always been pathetically unclear on the 30/360 day count. The book by Stigum and Robinson (the bond book, not the money market book) is a much better choice in my opinion. If you have Excel installed on your computer you can, however, see the relevant formulas in the built-in help file, just look for the financial functions there)
 
User avatar
dimitri
Topic Author
Posts: 0
Joined: July 14th, 2002, 3:00 am

Day count basis

June 16th, 2006, 8:14 am

Thank you very much it helped me a lot, although I have one more question. Do I calculate Accrued Interest in the same way in the Money market as in the Bond Marked?
 
User avatar
DavidJN
Posts: 269
Joined: July 14th, 2002, 3:00 am

Day count basis

June 16th, 2006, 2:38 pm

I'm not sure what you mean by your last question regarding the money market. Money market instruments (e.g. T-bills) generally do not pay coupons so there is no accrued interest per se. Perhaps you could cite a specific example?
 
User avatar
caroe
Posts: 2
Joined: July 14th, 2002, 3:00 am

Day count basis

June 19th, 2006, 9:06 am

There are several act/act conventions, so you need to be more specific, and this is also the cause of your problems. In bondland, act/act calculations are carried out according to ISMA rule 251, whereas in swapland the ISDA convention is commonly used. The attached ISDA memo will explain the differences to you
Attachments
Euro-swap-memo-ISDA-1998.zip
(54.95 KiB) Downloaded 74 times