Page 1 of 1
building a zero curve with lack of market data
Posted: June 2nd, 2015, 12:42 am
by arnold0715
This is Arnold Lin, a new learner in finance.I'm trying to build a 30-year CNY zero curve by China swap rate (for pricing a callable corporate bond).But I could only find the swap rate within 10 years.The only idea that comes out from me is using the Treasury data in China but it doesn't make sense. I want to ask if there is any better suggestion for building the 10-30 years interval.Many thanks!
building a zero curve with lack of market data
Posted: June 5th, 2015, 9:26 am
by studenttt
If you don't have instruments, you just can't build it, can you? Can you assume some kind of extrapolation?
building a zero curve with lack of market data
Posted: June 10th, 2015, 2:43 am
by arnold0715
Thanks for your replying.Yes, assumption of extrapolation is available.And I also wonder that in practice, which assumption of extrapolation is used commonly when building a curve like this.Many thanks again.
building a zero curve with lack of market data
Posted: June 11th, 2015, 8:12 am
by MHill
See an old discussion here:
http://wilmott.com/messageview.cfm?cati ... id=38700We had a similar issue a few years back, trying to get a 50y data point from a curve that went to 30y. We realised that the choice of extrapolation method could introduce volatility that should not be there (small change in inputs leading to large change in extrapolated values). We ended up just using the 30y value for the 50y value. For us, it was a small part of the portfolio, and we were doing risk management, not trading, so we were OK with the approximation.
building a zero curve with lack of market data
Posted: June 17th, 2015, 12:47 am
by arnold0715
Got it!Thanks for your answer and opinion.I'll think about it!