Dear Quantlib user group,
I would like to perform a surface of implied volatility matrix for equity Option and deduce from it a generic structure (i.e 6m,1y,2y,....) with bilinear interpollation. It looks like it's not yet developped in Quantlib. Is someone will be interested in debugging this developpement? (not yet done) I intend to copy the swaptionvolatilitymatrix to do something like Observable<-EquityOptionVolatilityStructure<-EquityOptionVolatilityMatrix Thanks Xavier |
Hi Xavier
>I would like to perform a surface of implied volatility matrix for equity >Option and deduce from it a generic structure (i.e 6m,1y,2y,....) with >bilinear interpollation. >It looks like it's not yet developped in Quantlib. It is not in the 0.3.0 release, but I've added it to the CVS in the last week. I'm not going to release a CVS snapshot in the next few days, but you can take a look at the development checking out the code using CVS (quantlib.org/cvs.html) For a quick glimpse follow these links (watch out the wrap): the vol term structure interface: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/quantlib/QuantLib/ql/voltermstructure.hpp?rev=HEAD&content-type=text/vnd.viewcvs-markup a constant vol term structure: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/quantlib/QuantLib/ql/Volatilities/constantvol.hpp?rev=HEAD&content-type=text/vnd.viewcvs-markup a time dependent vol curve: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/quantlib/QuantLib/ql/Volatilities/blackvariancecurve.hpp?rev=HEAD&content-type=text/vnd.viewcvs-markup a time/strike dependent vol surface: http://cvs.sourceforge.net/cgi-bin/viewcvs.cgi/quantlib/QuantLib/ql/Volatilities/blackvariancesurface.hpp?rev=HEAD&content-type=text/vnd.viewcvs-markup the classes are template of the interpolation method (as of now bilinear is the only 2D interpolation method in QuantLib), and use interpolation on variance instead of volatility. These functionalities are also available in QuantLibXL: to compile QuantLibXL you will need up-to-date versions of both QuantLib and xlw. In QuantLib Excel there is also a spreadsheet that explain why one should interpolate on variances and not on vols Early feedback on this work is more than welcome. I'm tackling these issues as side effect of a general review of the Pricing Engine framework Luigi suggested (see the original proposal at http://quantlib.org/quep/quep005.html) I pushed forward the engine framework, developing generic quanto and forward engines that will provide quanto/forward version of whatever existing engine you provide as input. The development line is to integrate in the pricing framework the generic class for Exercise (European, American, Bermudan -- Sad's suggestion) and last but not least to have all pricing engines able to handle time-dependent yield and vol curves. For vol surface we will need extraction of local volatilities from Black vol surface, and this won't be a quick job. Any volunteer for this task? Any help/feedback appreciated ciao -- Nando |
Free forum by Nabble | Edit this page |