"High performance CDO pricing and risk management systems can not afford to leave any available performance on the table," said Seyi Olurotimi, Research Head at Adaptive Data Engines, which operates fastmontecarlo.com, "CDOSampler integrates very siimply and seamlessly with current or proposed Monte Carlo risk implementations to deliver a sharp quality upgrade to the implementation. Similarly, ratings methodologies can be plagued by low accuracy or stringent performance penalties because they effectively try to calculate high accuracy estimates from low-sample tail regions of the loss distribution. CDOSampler again decisively solves this problem, offering enhancements to rating agency and other methodologies, and delivering crisp performance upgrades to those tools.
CDOSampler is suitable for enhancing CDO, CDO squared, and CDO cubed implementations. It is compatible with the normal and Student's t copulas, and admits single as well as multifactor correlation structures.