<t>Finite element method would be great for that as you can mesh the point of interest with needed accuracy.MATLAB pde toolbox should be able to do this, although it's been a while since I used it.I simulate similar problems in different field using COMSOL.For 1D as you have you should be able to co...

Thanks, Alan!I found couple good refs, in particular book excerpts from "Option Valuation under Stochastic Volatility".whoever wrote it, the book looks good

Guys,Can you advice on reference of non-stochastic PDEs for any type of pricing?So BS is equivalent to heat equation; Heston has a clean PDE form and an analytic solution.Are there nonlinear, nonstandard, nontrivial models (but still practical) out there?Thanks!

Wow! that is a little counter-intuitive!

Wow! so it is Itō-Doeblin Theorem now instead of Ito's lemma

<r>I had the same question and tried to write up a summary:Try this.<URL url="http://www.wilmott.com/messageview.cfm?catid=8&threadid=72684&FTVAR_MSGDBTABLE="><LINK_TEXT text="http://www.wilmott.com/messageview.cfm? ... SGDBTABLE=">http://www.wilmott.com/messageview.cfm?catid=8&threadid=...

actually the have a simple put example.provided the code is flexible in terms of boundary conditions it will be easy to price some weird payout calls.

Did you look into comsol?It seems to be rather flexible and has a pde module, where you can specify an equation yourself (within a certain pool of possible equations).

<t>QuoteWell, we can very well have a bunch of government employees running around moonlighting.you have your point, I guess.however without that opportunity government scientist does not have much motivation to work hard.Quotecool. if these C students are paid the same as your tough pde job, why do...

<t>Here are my two cents. I have a list of points. I spent 5+ years in one of these labs.Also note that the list contains things that bug me.The labs are very big and the situation varies across the lab.Some things you get used to and find workarounds.Some groups don't have many of the downsides, th...

Sorry, I don't know the rankings.You probably should have figured the question out before you enrolled.

The reason for displacing to (i-0.5)/n is simply because for many distributions invF(1)=infinity

I am a math background person who is interested in learning finance and would love to volunteer to read good books!

that's what I liked about Ginzburg.His initiative was - a regular review journal of debunking pseudoscience.They covered a broad spectrum of topics.It is a great read, unfortunately in russian.

Yep. I've done that now.Everything is clear, thank you!