I sorted out this (somehow). Key is the copounding rule: qlDiscount()
should use the Simple type in order to generate the argument for the
qlDiscountCurve() while qlForwardRate() has to use the Continuous one
generating the args for qlForwardCurve(). I'm not terribly sure I can
explain this but the results of qlParRate() using the two yield term
structures created above return with a reasonable approximation
qlPiecewiseYieldCurve() arguments.
Matteo
________________________________
From: Matteo Castagna
Sent: 20 December 2006 20:25
To: '
[hidden email]'
Subject: [quantlibXL] yield curves testing
I'm trying to learn how to use several functions related to interest
rate products. In order to do so I'm creating a yield curve structure
using the qlPiecewiseYieldCurve() function, from that structure I can
derive qlDiscount() and qlForwardRate() vectors. I'm trying to use those
discount factors and forward rates in order to create new interest rate
structures and with those get the qlParRate() for maturities I used
under the very first step.
All ok with the curve I define using qlDiscount() factors: qlParRates()
are very very close to the original input. Unfortunately it's not so for
the par rates I get using the curve defined using the forward rates. I
already checked several times the options I'm using (DayCount and
CompoudingRule) just in case I made some mistake there but it looks like
there's none.
Any help is welcome.
Matteo
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://sourceforge.net/mailarchive/forum.php?forum=quantlib-users/attachments/20061221/4801f35d/attachment.html
Locked
