Financial Modelling in Python

P1: JYS

c09 JWBK378-Fletcher May 12, 2009 19:0 Printer: Yet to come

128 Financial Modelling in Python

underlying = ctr.retrieve symbol("underlying")

underlying *= 0

cnt=0

for l in self.

trade.legs():

underlying += ctr.retrieve

symbol("leg"+str(cnt))

cnt += 1

ctr.update

symbol("underlying", underlying, to )

pv = 0

if self.

trade.has exercise schedule():

if self.

trade.exercise type() == exercise type.callable:

pv = ctr.retrieve

symbol("berm").mean()

else:

pv = ctr.retrieve

symbol("underlying").mean() \

+ ctr.retrieve

symbol("berm").mean()

else:

pv = ctr.retrieve

symbol("underlying").mean()

return pv

It needs to be emphasised that in the above pricing framework there is theoretically no

restriction on the dimensionality of the model. Everything will flow through just as well for a

multidimensional model, albeit more slowly, as it would do for a one-dimensional model. The

only assumption being made on the underlying numerical container is that it has the arithmetic

operators overloaded and a mean operator.

To conclude this section, we note that sometimes it is useful for diagnostics purposes

to be able to listen to the symbols at various stages of the calculation. The optional

symbol

table listener argument in the constructor provides this functionality. A con-

crete example, provided in the test module ppf.test.test lattice pricer,isgiven

below:

class european symbol table listener:

def

init (self):

self.

symbols = []

def