[Quantlib-dev] Vanilla dividend engines

[Joseph W. <jo...@co...>]

The place to start is to look at the

QuantLib/ql/PricingEngines/Vanilla directory in the CVS tree.  Take a 
look at the fddividendengine classes.  They calculate the option price 
using the Black-Scholes equation with explicit calculation for the 
dividends.  Right now I'm trying to figure out how to generalize the 
engines to be useful for things other than options (namely convertible 
bonds).

As far as how dividends are represented.  The are currently stored as a 
vector inside of the VanillaOption class.  This is suboptimal for a 
number of reasons.  The first is that it doesn't use the scheduling and 
calendar classes or have provision for different times of options.  The 
other problem that it bothers me is that the dividends should be 
assocated with a stock, and not the option.  This is going to be a 
problem with modelling convertible bonds since the bond could turn into 
a dividend-paying stock.

One other thing I would advise you to do is to be bold in refactoring 
the classes.  I like the basic design of quantlib, but there are a lot 
of things that could be cleaned up, which haven't been for lack of time.