Big Analytics for Financial Services

Daily trading operations generated massive volumes of data long before ‘big data’ became a buzzword and an Economist cover story. The Financial Services industry was one of the earliest communities to leverage scientific computing software and the quantitative skills of scientists for applications such as algorithmic trading and risk analysis.

Paradigm4, an advanced analytics database designed by scientists, brings new levels of scalability and functionality, along with a more cost-effective solution for high performance computing, to the financial services industry.

Feature Benefit
Scale up Complex Analytics
 Run complex math functions like covariance, regressions, PCA, SVD on matrices too large to fit in memory on a single node.
Time-Series Data
 Built-in support for time-series data without expensive or custom databases.
Support for Audits and Reproducibility
 Paradigm4’s no-overwrite data management is critical for reproducing results or rerunning simulations with the exact data used at the time. Keep all the original data and updated data needed to support audits.
Provenance
 Drill down to reconstruct the data that contributed to a result. Keep all the raw data, the derived data, and the derivation.
Uncertain Data
 Paradigm4’s support for user-defined uncertain data types embeds attributes with each data point, such as associated interest rate ranges, foreign currency exchange rate ranges, or probability information.
Missing Data Reason Codes
 Paradigm4 supports multiple, custom-defined null values (such as missing datapoint, market closed, trading halted, or feed unavailable) so that applications can substitute context-sensitive values.
 

Comments are closed.