Re: [Quantlib-users] Missing USDLibor6M Actual/360 fixing

[Luigi B. <lui...@gm...>]

On Wed, 2010-08-11 at 04:20 +0000, MonkeyMan wrote:
> Okay, one more follow up about this. What about instances like the following.
> Base currency is USD.
> 
> Eval Date: May 2, 2005 (Good day in US, holiday in UK)
> Settle Date: May 4, 2005 (Good day in US & UK)
> 
> Since it is a holiday in the UK, I need to use libor fixings for the previous
> good day, which is April 29. Luckily I have this data. Now I get this error:
> 
> 1st iteration: failed at 1st instrument, maturity May 11th, 2005, reference date
> May 4th, 2005: Missing no-fix1W Actual/360 fixing for April 29th, 2005

The bootstrap process is not using the fixings on purpose (the no-fix1W
tag is a dark hint.)  If it used them, the solver inside the
bootstrapper couldn't run as modifying the curve node would have no
effect.  I'm not sure how we can get out of this without modifying some
code.  To make this case work, you'd have to patch
DepositRateHelper::impliedQuote() and have it calculate the deposit rate
off the curve instead of using iborIndex_->fixing---or maybe you can
patch it so that the fixing date is moved to today.  I'm not sure which
is the least evil.

Luigi


-- 

No, I'm not interested in developing a powerful brain. All I'm after 
is just a mediocre brain, something like the president of American 
Telephone and Telegraph Company. 
-- Alan Turing on the possibilities of a thinking machine, 1943.