Hi,I am looking for a general pricing mechanism to price a single and/or double window barrier option.A single barrier has only 1 KI ot KO level and a window period. A double barrier has a KI or KO lower barrier with a lower window period. It also has a upper barrier with a upper window period. (Note: the two window period may be overlapping)Thank you.

I think you're going to have to resort to numerical pricing for such a beast. Finite difference methods are the standard.

No close form is possible. Chee, give the exact specification ...

A window barrier option will specify the barrier's active period. That is the only difference from a normal barrier option.Other as per barrier option.

QuoteOriginally posted by: pb273No close form is possible. .>I think you're going to have to resort to numerical pricing for such a beast. Finite difference methods are the standard. __________________________________________________________Artur Sepp and I are working on a paper "Option Pricing with Jumps".Our main results relate to jump diffusion, jump volatility and so on.The question about window barrier option (also kind of a jump process) inspired us to add a short section about "such a beast".Here is an extract from the early version of our paper that can help to understand how to treat those problems analytically.

Window Barrier with alternating Low, High barriersThey are correct in saying that there is no closed form solution. usually you do this with numerical integration type models i.e.1. At maturity you generate an up down payoff vector (how far up down depends on the presence of barrier at this point.2. For each event point going backwards, you assume that there are double barriers, if there is only one 'real' barrier, you set the other to 6+ standard deviations from the drift adjusted spot price at that point.3. At that event point the up/down vector of values is from barrier to barrier, but one barrier is not real.4. You calculate the intermediate value for each spot point at that event point5. You step back through each event point until you get to the start date, at which point the price is the integral of all the valuesat the previous event point ( start or end of first barrier).it s complicated , but then it's a complicated model