Re: Re: MCMC in Mathematica

*To*: mathgroup at smc.vnet.net*Subject*: [mg96091] Re: [mg96043] Re: MCMC in Mathematica*From*: Andrzej Kozlowski <akoz at mimuw.edu.pl>*Date*: Wed, 4 Feb 2009 05:23:18 -0500 (EST)*References*: <gm0q4e$rmq$1@smc.vnet.net> <200902010943.EAA22903@smc.vnet.net> <gm6kti$a11$1@smc.vnet.net> <200902031131.GAA00243@smc.vnet.net>

On 3 Feb 2009, at 12:31, Jens-Peer Kuska wrote: > While the most things done rather quick in Mathematica, a simulation > can run several days and so you may run into the case, that you > miss the license to answer the Mathematica newsgroup because you have > a Mathematica 6 and a Mathematica 7 open, to check your answer with > both versions ... But there are simulations and simulations. I run a lot of Monte Carlo simulations that compute prices of various kinds of financial options in Mathematica. They take maybe two minutes to run and give answers accurate to 2 decimal places, which is probably better than the accuracy of the models. Even if it would take only a second to run this in C, the programing time would be vastly longer (even assuming that you have access to a library of mathematical and statistical functions that you can use in your C program). (Besides, at least in the case of option pricing, it is a terrible idea to rely on only one method (e.g. simulation) to obtain the answer, and Mathematica let's you use all know ones with minimum of programming). So the answer to the OP question depends crucially on what sort of simulations he has in mind. Andrzej Kozlowski

**References**:**Re: MCMC in Mathematica***From:*Jens-Peer Kuska <kuska@informatik.uni-leipzig.de>

**Re: MCMC in Mathematica***From:*Jens-Peer Kuska <kuska@informatik.uni-leipzig.de>