Bug in implied volatility - please read

classic Classic list List threaded Threaded
1 message Options
Reply | Threaded
Open this post in threaded view
|

Bug in implied volatility - please read

Luigi Ballabio-2
Hi all,
        I just fixed in CVS a serious bug in the impliedVolatility()  
method of all classes derived from OneAssetOption (you can check the  
documentation to see the affected classes) which slipped through the  
test suite and could cause wrong results to be returned if a certain  
sequence of actions were performed. We might consider releasing  
QuantLib 0.3.6 sooner than we thought. In the meantime, here is the way  
to avoid triggering the bug:

a) if you mean to call the impliedVolatility() method on such an  
option, do not pass it a Handle<StochasticProcess> you already passed  
(or intend to pass) to another instrument. Create a new Handle  
containing a newly-allocated process.
b) impliedVolatility() will return the right result. However, after  
calling impliedVolatility(), the option is broken. You might as well  
throw it away and instantiate a new one.

Sorry for the inconvenience,
                                Luigi