April 13th, 2006, 1:59 pm
Experts -As I understand, the BGM model is fundamentally defined under the terminal measure, and therefore to price a contingent claim using a monte carlo framework, we have to transform the payoffs to their terminal values before averaging over the various simulations (ie whatever the opposite of "discounting" is in order to get the "Net Terminal Value").My question is this (assuming the above is correct): if we have a payoff H at time Tk and need to transform it to its terminal value at time TN, do we do the transformation using:* the LIBOR rates as per the forward curve as simulated at time Tk (e.g. the value of Lm that the simulation generates when t=Tk, k<m<=N)or * the "final" LIBOR rates for the simulation (e.g. the value of Lm that the simulation generates when t=Tm, k<m<=N)?Thanks!