Hull uses continuous compounding for the yield in his bond pricing examples. In the sample question P1.T3.170.3 we are asked to compute the dirty price of a bond where the yield is given an 6% without any explicit compounding interval assumption. The solution uses a calculator to compute the price - thereby implicitly assuming that the 6% yield is quoted with semi-annual compounding The question I have is should we always assume that the yield quoted with reference to a bond in the exam is with semi-annual compounding if the compounding interval is not explicitly stated? That is either use semi-annual directly of convert the yield into the continuous equivalent as proceed in a "Hull" fashion (in this example the continuous equivalent yield would be 5.912%
Default compounding assumption for yield in bond pricing questions
- Thread starter neveo
- Start date
Hi @neveo You might notice this has been frequently discussed (no pun intended! 
), or at least a variation on the compound frequency issue. The FRM will specify, as GARP is highly aware of the fact that various frequencies can be employed. You are correct Hull defaults (no pun intended again!) to continuous, but the exam (if history is a guide) can use either continuous or discrete. I will say that the only unlikely scenario is a need to do compound frequency translations: while those are good for learning/fluency, the FRM does not really embed the need to translate. Either you'll get an explicit states assumption of continuous or discrete (and if discrete, then likely annual or semi-annual; annual is common due to exam ime constraints). I hope that helps,
), or at least a variation on the compound frequency issue. The FRM will specify, as GARP is highly aware of the fact that various frequencies can be employed. You are correct Hull defaults (no pun intended again!) to continuous, but the exam (if history is a guide) can use either continuous or discrete. I will say that the only unlikely scenario is a need to do compound frequency translations: while those are good for learning/fluency, the FRM does not really embed the need to translate. Either you'll get an explicit states assumption of continuous or discrete (and if discrete, then likely annual or semi-annual; annual is common due to exam ime constraints). I hope that helps,Similar threads
