The STCDO (single-tranche collateralized debt obligation) Pricing Engine employs a Gaussian copula framework with stochastic recovery, semi-analytic method, including an internal credit default swap (CDS) calibration to extract the hazard rates from CDS quotes and a CDO calibration to extract base correlations from a standard credit index (e.g., DJ Itraxx).
The STCDO Pricing Engine prices standard credit index tranches and includes the following features:
- Arbitrary number of names in the CDO and their corresponding portfolio weights.
- CDS spreads (or coupons) and upfront points for each name in the CDO.
- Discrete levels of stochastic recovery for each name and the corresponding probability level.
- Coupon date and maturity date adjustments, roll conventions, choice of day count basis and holiday calendar support.
- GPU-enabled— takes advantage of NVIDIA graphics processors and delivers much faster accelerations than serial code. No CUDA or parallel programming expertise needed.
Single-tranche CDO Pricing Engine Output Data
- Tranche present value
- Tranche fair spread, given the upfront points
- Tranche fair upfront point, given the coupon rate
- Tranche mark-to-market value (holder)
- Tranche cash settlement amount (buyer)
Like all SciComp Consulting solutions the STCDO Pricing Engine can be enhanced/modified to meet any particular modeling needs you may have and is available as C/C++ source code, Windows/Linux executable, or a ready-to-use Excel spreadsheet and add-in.
The STCDO Pricing Engine can be enhanced to support the modeling of bespoke STCDOs with a number of approaches for bespoke portfolio mapping of the base correlation available, including:
- Using the same BC curve as for the standard portfolio
- ATM method
- Probability matching method
- Tranche loss proportion method