QuoteOriginally posted by: odradekYou mean an MC cluster? That´s awesome. Asked about something like that here but we was told something about having limited resources. Be lucky if they give me another machine next year (though I guess that would technically be a cluster, population 2!) So until then looks like I´ll be spending a lot of time with MC, myself and I.Not really a cluster I'm afraid, I'm not good enough at the networking type techy stuff (or have the time to bother) to make it into a proper cluster. I have 2 4CPU servers and 1 2 CPU server, so can run 10 threads at once without my desktop, and generally just do this by running multiple instances of the same code.

QuoteAren't baskets just F(4) Hamiltonians?You lost me a bit there. Been working on methods for valuing baskets myself.Is this related to a differential geometry approach (solving heat equation on a Riemannian surface)?-OD

QuoteBeen working on methods for valuing baskets myself.Is this related to a differential geometry approach ?-ODIf I may intrude here. One way to view baskets is a PDE with an elliptic quadratic form. These forms then have a geometrical interpretation but in N dimensions it's difficult to visualise.Numerical schemes (e.g. FDM) should preserve ellipticity to ensure monotonicity. BTW which methods are u using for baskets?

QuoteSo until then looks like I´ll be spending a lot of time with MC, myself and I.Sounds awful lonely BTW what do you think of these performance results?Computed it with 80000 runs on MC2 factor -> 8 sec.3 factor -> 14 sec.4 factor -> 19 sec.

MC I run in the bacground in Xcode, get a life get a Mac!

QuoteBTW which methods are u using for baskets? I´m currently testing out a Riemann manifold approach (similar approachto an Avellaneda paper from a few years back, but with time-dependencyin the vols) to get an asymptotic expression for basket volatility, using local vol MC to test the results. QuoteMC I run in the bacground in Xcode, get a life get a Mac!If I had my druthers, I´d be doing my thang in Linux. But we mustuse what the rest of the herd uses, mustn´t we?moo.

QuoteI´m currently testing out a Riemann manifold approach (similar approachto an Avellaneda paper from a few years back, but with time-dependencyin the vols) to get an asymptotic expression for basket volatility, using local vol MC to test the results.Do you have a reference to that paper? thx.

Quote Do you have a reference to that paper? thx.It´s available on his NYU page here

We have a cluster and a task engine, etc. It's very nice to send the request off with a level of importance and to get an email back with the results.