Hello Evereyone!I am pricing barrier options analytically as well as using Monte Carlo simulation in Matlab(GUI). I am getting a big difference between both the option prices. Can anyone tell me how to minmize this difference?Thanks.Festivie

Hi!Well, so in general the only thing I can think of is just to increase the number of simulations... how many have you preformed?but maybe if you give some more details...bye

In the case of barrier options, when you discretize, you are omitting the proba that the brownian crosses the barrier between the discretization points, so you should first think about decreasing the time steps to see if it converges closer to the closed form formula. If not that means you probably have some error. Otherwise, you need to use brownian bridge to correct for the problem.

Hi,Brownian Bridge techniques are the key. You can also find interesting stuff under theterm "Brownian Interpolation". Just put it into google...Have fun

see my recent paper on www.markjoshi.com

Hi everyone,I should be very interested in implementing the BB stuff, for the barrier option; not sure however, with the way to practically carry out the matter! Should someone help, plz?I propose:1: simulation of n paths for the underlying asset2: identify for each path, a time where the spot is closed from the barrier, say t_i1 is this time3: enrich the path with simulating a Brownian Bridge between t_i1 and t_i2 (i2=i1+1)Is the process correct, or am I wrong? Cheers.

If you are found of MC I think you would be intreasted to look into my report on an generall MC implementation.RegardsChristergenerallMC

When dealing with Monte Carlo and error reduction. You should try to utilze an error minimization technique such as control variate etc.

Hello,The link does not seem to work as far as I am concerned. Do you confirm?Thanks.

Just remove the https:// in front and it works!

QuoteOriginally posted by: md98colIf you are found of MC I think you would be intreasted to look into my report on an generall MC implementation.RegardsChristergenerallMCI am not only fond of MC and its versions but find it very levelling. It is a poor quant's tool to solutions which might otherwise require a lot of brain taxing and revisiting the highschool(graduate sometimes) text books. Anyways, just went through your thesis briefly. Seems,AWESOME. great work done. Congratulations. Cann't say much about the technicalities. Cuch should be the right person to say about the software part.cheers IMPORTANT DISCLOSUREzoloren, learn to be attentive. After you click on the link,you should see what happens to the address. There cann't be multiple https in this world.Patience is the key, to every important aspects of life.