Typical SciFinance Pricing Model Development Projects

Implement proprietary trading and arbitrage strategies

Customers seeking to shield the proprietary nature of their trading strategies through internal model development find that SciFinance leverages in-house developers by an order of magnitude. SciFinance allows quants to focus on trading strategies rather than spending time with rote programming of models. And because developers can make all of the modeling decisions, no two groups are likely to produce the same model.

Pricing Derivatives

Hedge risk exposure

Customers looking to implement market standard pricing models for use with their portfolio/risk management systems find that SciFinance provides a cost effective, easy to integrate, transparent and flexible modeling architecture.

SciFinance comes with hundreds of pricing model examples across asset classes that users can tailor to meet their particular needs. Modeling decisions such as the choice of underlying dynamics, which market data to use and its format, and what sensitivity measures to compute are all under a customer's complete control.

The SciIntegrator™ module automatically generates wrapper code and associated projects and test frameworks for any SciFinance-generated pricing code, facilitating the testing and integration of pricing models with in-house or third party applications.

SciFinance Customer Overview

>Download the SciFinance Customer Overview

Validate pricing models

Customers tasked with internal model validation responsibilities find that SciFinance provides an optimal solution for comprehensive pricing model validation. SciFinance allows rapid validation of existing models via alternative techniques. Changing a few keywords modifies the underlying model dynamics, numerical methods, solvers, finite difference schemes, Monte Carlo discretizations, and/or sensitivity measures.

In the vast majority of validation tasks where the dynamics are of low dimension, Monte Carlo models can be validated against PDEs and visa versa in minutes.

SciFinance provides a quantitative analyst full control over the implementation of the pricing model and complete model transparency. SciFinance contains no "black box" components. All pricing model libraries included in SciFinance are provided as source code libraries and are open to customer review and inspection.

Value and risk manage derivatives

Customers looking to offer derivatives valuation, portfolio management and other risk management services find that SciFinance provides comprehensive asset class support for implementing their derivatives pricing models. With access to hundreds of customizable, industry-proven examples as a starting point, a single quantitative analyst can create custom models in minutes.

SciFinance provides robust calibration tools for many industry-popular pricing models. In addition, a model calibration development framework is available for customers who need to implement custom calibration functions.

Optional components of SciFinance automatically generate parallel code (CUDA or OpenMP) for any Monte Carlo pricing models. PDE pricing models can be GPU-enabled on a consulting basis. By adding the keyword "CUDA" or "OpenMP" to an existing single threaded model specification, SciFinance automatically generates parallelized pricing model source code that can be run on NVIDIA GPUs or multi-processor machines CPUs. No CUDA or parallel programming skills are required.

SciFinance-generated code can be seamlessly integrated with trading and risk systems. The process is made completely automatic by the SciIntegrator™ module. By simply adding a keyword to the model specification, customers can automatically generate Microsoft Excel, Java, .Net, COM or Python interfaces (customized interfaces are also available). SciFinance does not impose proprietary data models to hinder code integration by requiring wasteful data container translations.

Need more information on SciFinance and automated derivatives pricing model development?
Contact us >>