Serving the Quantitative Finance Community

 
User avatar
quanteric
Topic Author
Posts: 6
Joined: June 4th, 2010, 12:01 am

USD curve trades

June 12th, 2018, 10:55 am

Hi everyone, am wondering if you can help me with a little problem. I was using Bloomberg USD curves to imply out the 2-5-10 butterfly by valuing the individual swaps and took 2*5y-2y-10y as the quote. This when compared to other banks was around 3 bps out, which is higher than I expected. I tried turing to OIS discounting, but it made only a slight difference.

I guess it may be due to ... 1) interpolation, 2) choice of instruments (3m libor than next 12 imm eurodollars, then swaps..) or 3) convexity asjustment.

Does anyone know if there is a standard market practice for building the USD curve? Thanks very much for your help on this.
 
frolloos
Posts: 752
Joined: September 27th, 2007, 5:29 pm
Location: Netherlands

Re: USD curve trades

June 12th, 2018, 11:22 am

I am not an IRD expert but in my experience the pricing source and timing/time zones can make a difference (assuming the curve construction is in line / you are looking at standard maturities). Are you using ICAP as pricing source? 

Using OIS discounting is more or less standard these days, so you probably want that "on" at all times.
 
User avatar
Martinghoul
Posts: 188
Joined: July 18th, 2006, 5:49 am

Re: USD curve trades

June 17th, 2018, 10:22 am

BBG USD curves are normally reasonably accurate, so you really shouldn't have been out by 3bps on the fly.  It's very hard, however, to pinpoint the potential reasons for this, as it could be a bunch of things.
 
User avatar
quanteric
Topic Author
Posts: 6
Joined: June 4th, 2010, 12:01 am

Re: USD curve trades

June 19th, 2018, 8:37 am

Thank you very much for your reply Martinghoul.  I am wondering if I can pick your brain further...

1. What would be an acceptable error for the butterfly?  Within 1 bp?  Or less?
2. Currently I am building the USD curve with 3M Libor, then the next twelve HMUZ Eurodollar futures, and then swap rates from 4y onwards.  If I compare this against the quotes for IMM swaps, I am out by 2 bps or so on the 2y Dec-Dec IMM (cob 11 Jun 2018).  Do you think it is reasonable if I put in the 2y IMMs as curve-building instruments instead of using Eurodollar futures between 3y and 4y?

Thanks very much for your help on this.
 
User avatar
Martinghoul
Posts: 188
Joined: July 18th, 2006, 5:49 am

Re: USD curve trades

June 27th, 2018, 8:30 pm

If you're getting your prices from BBG, you should be within 0.5bp, I would think.  Hard to know, though.

Have you checked the convexity adjustment you're applying to your ED futures?  It's reasonable to do whatever you want to do, but normally people prefer to use futures, rather than swaps for the short end.