Numerix, a provider of cross-asset analytics for derivatives valuations and risk management, today announces expanded real world modeling coverage in the Numerix CrossAsset core analytics stack, including the availability of the new Hull-White 2-factor real world interest rate model.
“Numerix is unrivaled with the deepest, most flexible risk-neutral capabilities in the industry. However, to ensure the most complete and comprehensive coverage for producing consistent risk-neutral and real world economic scenarios, over the past several months we’ve focused on expanding our real world modeling coverage,” said Steven R. O’Hanlon, Chief Executive Officer & President of Numerix. “By enhancing our sophisticated stochastic simulation framework we are better positioned to empower our clients to meet the regulatory requirements set forth in Solvency II, IFRS, GAAP, AG43 and C-3 Phase II.”
Addressing Real World Modeling Challenges
“Years ago in the insurance space there was very little stochastic modeling happening. While mortality and longevity risk were well understood – how the liability was behaving when guarantees and riders were added on top of base contracts introduced new types of risk,” said Pawel Konieczny, PhD, Vice President, Insurance Solutions, Client Solutions Group at Numerix. “Today it’s clear that the guarantees insurance companies are selling, like riders on variable annuities behave like derivatives and have to be managed like derivatives. As such leveraging a core set of real world models within a stochastic modeling framework for things like economic capital and hedging is critical, especially when looking to generate nested stochastic simulations.”
While insurance institutions have been a major driver in furthering Numerix’s real world modeling capabilities, there are many varied uses cases for real world scenarios. Some examples being economic capital, product pricing, enterprise risk management, variable annuity capital and reserves, strategic asset allocation and asset/portfolio management. Numerix’s enhanced coverage applies broadly to pension funds, asset managers and portfolio managers – or any user that requires real world Monte Carlo analysis of their portfolio. This can extend to determining value at risk, stress testing or optimizing allocation of assets for minimizing risk or maximizing return.
Numerix Real World Modeling Coverage
Numerix has introduced the Hull-White 2-factor real world interest rate model (HW2F RW) which supports calibration to a projection curve and historical time series data. Users interested in building a real world calibration for calculating economic capital, fitting defined capital market expectations for setting up a strategic asset allocation, or a portfolio optimization can utilize the HW2F model. The Cox-Ingersoll-Ross 2-factor real-world interest rate model (CIR 2F) is also available, calibrated to a projection curve.
For commodity coverage, the real world version of the Black commodity model can be calibrated to time-series. And for equity and FX coverage calibration of the Bates real world model to historical time-series is possible. Calibrating real world correlations to time-series data in the hybrid model has also been added.
Economic Capital: Model Flexibility and Speed
In the case of a calibration for economic capital, some insurance companies aim to have a set of stochastic rates hit target rates over a future time horizon. Around each point in time, insurers would typically also have a volatility target, or a best estimate to hit.
In leveraging the CIR 2F model, all parameters may be time varying, including volatility related parameters. This allows users to achieve a desired distribution of interest rates that covers the entire projection horizon. Users are able to observe the distribution of real world interest rates at each point of time in the future using a single calibration.
“The CIR 2F model is affine which allows a lot of quantitative short cuts enabling a much faster computation of simple quantities like zero coupon bond prices. In thinking through some of these challenges before coming to market, Numerix is able to bring enormous benefit to its clients. Parameters must be flexible enough to fit calibration targets, and by using a CIR 2F model we offer enough flexibility to control the shape of the interest rate distribution through time. Fast calibration and simulation of the model due to its affine nature enables speedier evaluation of scenarios over a non-affine model such as Black-Karasinski,” adds Daniel Schobel, Actuary in the Client Solutions Group at Numerix.
In addition to being more efficient in terms of speed, Numerix’s approach gives users significantly more flexibility in terms of the resulting distribution of scenarios that can be generated out of the model. This is what enables Numerix to provide custom real world calibrations for a company’s economic capital or other use case needs.
“While real world model selection and calibration is clearly important, the best choices depend heavily on the specific use case, as one size fits all is rarely appropriate. That said, the interaction between scenarios and the systems that consume them is critical, and understanding how calibration choices may impact final results is essential,” said Schobel. “With Numerix’s ESG users have complete flexibility and transparency to view and customize all model and calibration settings. In giving users full control over scenario generation and enhancing our real world coverage we’re giving client’s even more tools to construct an appropriate real world ESG.”