A broker just quoted me a OAS with a volatilty assumption of 14. I asked him how he came up with that assumption and he said it is the street’s standard. Can anyone confirm this?
I will confirm it for 20% commission.
Yeah its the standard assumption. I can’t exactly remember why, but an old lecturer of mine stated so.
Seems right. From google… [PDF] FundingNotes: Wide Spreads of Fannie Mae Callables Present … File Format: PDF/Adobe Acrobat - Quick View and also relative to Fannie Mae bullet curve. The OAS relative to Treasuries was calculated using a fixed 14% volatility assumption, while the OAS relative … www.fanniemae.com/markets/debt/pdf/fundingnotes_11_98.pdf;jsessionid… [PDF] Fixed Income Weekly Market Outlook United States File Format: PDF/Adobe Acrobat - Quick View OAS UST 14% represents the option-adjusted spread of the callable bond to the US. Treasury curve under a 14% fixed volatility assumption. … pluto.mscc.huji.ac.il/~mswiener/teaching/…/FIWeekly.pdf - Similar
Can I ask a dumb question. How exactly do you use the vol assumption? (I guess I don’t know how OAS is calculated.)
Use the vol assumption to compute the value of a call (or put) option on a bond (I don’t remember exactly how to do this, but a net search should find some formulas.). Then. 1. Compute the zero volatility spread of the bond with the option removed. i.e. add the call premium to the current bond price, or subtract the put premium (remember you sell a call with callable bonds, but you buy a put). I believe the OAS is the number you get from step 1 (but someone can correct me if I’m wrong). Then you might… 2. Compute the zero volatility spread of the bond, assuming its current price is correct and there are no. 3. The difference between the spreads represents the effect of the option on the yield. In general, adding a call will increase the yield… adding a put will decrease the yield. If the vol assumption changes, it will increase or diminish the effect… more vol => more change between the OAS and ordinary z-spread.
DarienHacker, you will learn more than you will want to about calculating OAS with volatilty assumptions in Level 2.
job71188 Wrote: ------------------------------------------------------- > DarienHacker, you will learn more than you will > want to about calculating OAS with volatilty > assumptions in Level 2. Too late! They must have added it since 2006, or I forgot it. My understanding (from Hull, who doesn’t always agree with the rest of the world) is that OAS is calculated like z spread. So you (a) ignore any optionality, (b) find the spread to treasuries that lets you match the market price. This is obviously quite different from bchad’s procedure - no need to do any option valuation at all. The latter seems to agree with http://www.investopedia.com/study-guide/cfa-exam/level-1/fixed-income/cfa42.asp
you shouldn’t “ignore any optionality” because then it wouldn’t be “option-adjusted” spread. you have a plain vanilla bond and a market price quote, say 106%. you can find what the implied discount rate is which reconciles the contractual cash flows to the quoted market price, and obviously the spread over treasury from there. now you have the same bond but with embedded option (say a call). the quoted market price is 102%. you can do the same exercise (use the contractual cash flows, find the discount rate that calibrates to the market price). because the quoted price is lower, you’ll find a higher spread over treasury. but it won’t factor the presence of the embedded call option which reduces the bond price, so it really won’t be comparable to the first one. so you want to use a model (say binomial) which takes into account the effect of this embedded call option on the price, and then find the implied spread over treasury. this will be your OAS which is then more comparable to the first one. it will depend on how you modeled the effect of the call presence on the bond price, hence it’s model-dependent.