R: Why is Quant math implemented in Floating Point?
Posted by
Gianni Piolanti on
URL: http://quantlib.414.s1.nabble.com/R-Why-is-Quant-math-implemented-in-Floating-Point-tp4929.html
Hi Peter,
As far as I know if you need to generate sobol numbers the algorithm is better implemented using
floating point arithmetic.
In my little contribution to developing a faure sequence (you can find it under randomnumbers directory)
we have only used long int data types.
In my opinion you should study every algorithm and try to "reverse" it to the
integer arithmetic that best fits your cpu but before doing it you should also pay attention to
integer versus floating operations benchmarks.
ciao,
Gianni.
----- Messaggio Originale -----
Da: Moreton, Peter
Inviato: 13 July 2006 12:20
Oggetto: [Quantlib-users] Why is Quant math implemented in Floating Point?
Hi,
I have a simple question, which I'm sure has a straightforward answer: Given that some of the commonly used Quant algorithms are quite compute intensive, especially the Monte-Carlo routines, why are they so frequently implemented using Floating point maths? - wouldn't these routines run much faster if coded using scaled-integers?
Peter Moreton
Getronics UK Ltd
-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
QuantLib-users mailing list
[hidden email]
https://lists.sourceforge.net/lists/listinfo/quantlib-users
Prendi il massimo dal Web. Scarica GRATIS MSN Explorer dal sito:
http://explorer.msn.it/intl.asp#it