User avatar
Posts: 1733
Joined: July 14th, 2002, 3:00 am

Basis swaps valuation

July 31st, 2013, 8:15 pm

All this stuff depends on whether the swaps you are valuing are collateralized. If the swaps are collateralized you have quite a bit more work to do. Specifically, for collateralized products in major currencies (USD/EUR/CHF etc.) you need to construct a dual curve solution ? an OIS curve for discounting purposes and a projection (i.e. LIBOR) curve for projecting forward rates (conditional on OIS discounting). How to treat uncollateralized swaps appears to be an unsolved problem, at least the last time I looked ? in this case many people seem to be using older single curve solution (i.e. same curve used to forecast forward rates and discount) even though they may be aware it might not be strictly correct.Since a basis exists between floating rates with different tenors, one should not mix different floating rate tenors when building curves. The vanilla curve against 3-month floating resets should not have any other floating rate tenors other than 3 months in it. This creates a very interesting problem as to how to build the very short end of the yield curve (e.g. how to seed the futures strip). Opinions vary on how to handle this challenging problem, an excellent example of how curve building has become art as well as science.You have a great deal of researching and reading to do to bring yourself up to speed with the changes in the swap market since the last financial crises. And because I know you will ask, imho there are not any particularly clear, intuitive papers on the topic and pretty much all textbooks on this subject are now out of date. Good luck!
User avatar
Posts: 640
Joined: January 24th, 2005, 8:08 am

Basis swaps valuation

October 20th, 2013, 8:18 am

JohnyEnglish, If you come back to this post and read David's post then read this one ,too.All that David said is correct. You need to have a book which will give you the changes that has happened ON THE vanilla construction method.BUT, i have a little doubt that you have not got a good book on Vanill methods yet which makes you to confuse bettwen 3m and 6m rates etc.Can I suggest a book for BASIC stuff which will make you smart on vanilla curve building. The book is by Flavell, swaps and other derivatives.If you have not gone through a book like that before, do yourself a favor, go through the entire book. i assume that you are young, you will have a good future with a massive confidence boost. I am teaching my 14 year old son about IRD,too.Then, come back and ask David about some papers by1. Bianchetti. 2 Akihiko Takahashi and Masaki Fuji 3. Henrard 4. Fabio Mercurio etc. These papers have happened in last 5 years or so.There was one paper long back in 2003 or 2005 which had pointed out the need of having a discounting curve different than forecasting curve. I did replicate that paper in excel but not sure in which office I left that excel copy(i don't remember how many jobs I have changed in last 10 years). cheers

PW by JB has been "Serving the Quantitative Finance Community" since 2001. Continued...

Twitter LinkedIn Instagram



Looking for a quant job, risk, algo trading,...? Browse jobs here...