nZetta Derivatives Pricing Toolkit
nZetta Toolkit C++17 components are precision engineered using high-performance algorithms and the latest approaches specific to finance industry needs. With the nZetta Toolkit, you can write code once and seamlessly execute it on various platforms, including x86 single-threaded, x86 thread-distributed, or GPU, without any need for modification. This flexibility allows you to effortlessly switch between technologies, empowering quants to harness the full potential of their code while staying at the forefront of innovation.
Includes generators, correlation & repair, Brownian-bridge, antithetic, tranching, skip-ahead, regression, reduction, fast processes.
The nZetta Monte Carlo achieves over 15× speed-up on single-threaded x86 and over 120× speed-up on low-end GPU using the same quant code.
A C-API drop-in replacement supporting Douglas-ADI, Craig-Sneyd, Modified Craig-Sneyd, and Hunsdorfer-Verwer methods along with necessary ancillary functions.
With the nZetta Toolkit you can easily investigate and choose the best method for your problem; for a given desired accuracy, Hunsdorfer-Verwer can be 30% faster than Modified Craig-Sneyd.
Interpolators, processes, volatility models, etc.
Full tensor mathematics, shaping, regressions, reductions, windowed reductions. Fully vectorize your pricing library.
The nZetta Toolkit is developed by Zettamatics, experts in financial modeling, and NAG, the world’s leading provider of advanced mathematical analytics. This partnership merges Zettamatics’ extensive decades-long experience in front-office banking with NAG’s pioneering analytics capabilities. The result is the nZetta Toolkit, a collection of state-of-the-art components for derivative pricing that not only offers unmatched speed but is backed up with a strong technical foundation you can rely on with utmost confidence.