|
|
Hi, I am a QuantLib newbie and am going over code by an ex-coworker that bootstraps a swap curve from a combination of MM deposits, Eurodollar futures, and swaps. Quantlib version is QuantLib-0.9.7
The code ultimately calls into IterativeBootstrap<Curve>::calculate() function to generate the discount factors (interpolation = linear).
Within calculate(),
try {
BootstrapError<Curve> error(ts_, instrument, i);
Real r = solver.solve(error,ts_->accuracy_,guess,min,max);
// redundant assignment (as it has been already performed
// by BootstrapError in solve procedure), but safe
ts_->data_[i] = r;
} catch (std::exception &e) {
calls into QuantLib::Solver1D<QuantLib::Brent>::solve, which causes a crash at one point.
The crash occurs when the solve function is called with --
double accuracy=1.0000000000000001e-015,
double guess=0.98669969631115340,
double xMin=2.2204460492503131e-016,
double xMax=0.98737707359787918
This causes both fxMin_ and fxMax_ to be positive, causing
QL_REQUIRE(fxMin_*fxMax_ < 0.0,
"root not bracketed: f["
<< xMin_ << "," << xMax_ << "] -> ["
<< std::scientific
<< fxMin_ << "," << fxMax_ << "]");
to fail.
I am having difficulty figuring out exactly why both fxMin_ and fxMax_ are turning positive at that point. Would greatly appreciate any insights into this. Please let me know if I could provide more details to help.
Thanks
|
Locked
