Financial Modelling in Python

P1: JYS

c08 JWBK378-Fletcher May 12, 2009 18:58 Printer: Yet to come

The Hull–White Model 121

... , duration = ppf.date time.years

... , shift

method = shift convention.modified following)

>>> structure = trade((pay

leg, receive leg), (ex sched,

exercise

type.callable))

>>> factory = hull

white lattice model factory()

>>> hwmodel = factory(structure, env)

The corresponding factory for the Hull–White Monte-Carlo model is shown below. The

details are almost identical to the factory for theHull–White lattice model. The main differences

are that the model argument dictionary is used to control the valuesof the number of simulations

and start seed, and instead of a rollback component being constructed we construct an evolve

component. We finish off by constructing an exercise component for the case when the trade

is callable.

class hull white monte carlo model factory:

def

call (self, trd, env, model args = None):

ccy = ppf.core.enforce

single currency(trd)

terminal

T = env.relative date(ppf.core.final important date(trd))/

365.0

num

sims = 1000

if model

args <> None and model args.has key("num sims"):

num

sims = model args["num sims"]

seed = 1234

if model

args <> None and model args.has key("seed"):

seed = model

args["seed"]

s = monte

carlo.state(num sims)

ev = monte

carlo.evolve(ccy, seed)

f = fill(terminal

T)

r = requestor()

ex = None

id = 0

if model

args <> None and \

model

args.has key("explanatory variables leg id"):

id = model ...