Introducing Unified Modeling on the Intelligent Risk Platform with Moody’s RMS and Nasdaq’s Risk Modelling Services
Cihan BiyikogluMay 10, 2023
What does it take to get a more complete picture of risk? For many, it’s a daily struggle, riddled with complications such as data preparation, data integration, and data enrichment - all of which drag along the many other systems that provide the overhead plumbing.
This means that obtaining a more complete picture of risk simply becomes too costly, and too arduous for an insurance field practitioner, impacting their business agility and market competitiveness, because of these reasons:
Costly and difficult to utilize multiple views of risk.
Costly and difficult to take the focus away from core functions (e.g. risk modeling, underwriting, portfolio management) and build internal data taxiing systems.
Costly and difficult to manage multiple disparate modeling systems that are too expensive and difficult to maintain.
All these reasons also result in slower innovation in the modeling market.
Today, we are excited to announce a range of new innovations that represent a leap forward to help ease these challenges.
With a new modeling architecture, the IRP now looks to unify Moody’s RMS, third-party, and custom risk models and also to simplify the complex infrastructure required to bring together multiple views of risk for smarter risk decisions.
Unifying Moody’s RMS, Third-Party, and Custom Models
On May 10, at the Moody’s RMS Exceedance 2023 conference, customers saw a preview of the Risk Modeler™ application running Moody’s RMS, third-party, and other custom models.
With this new capability, we will be able to bring together under the IRP all Moody’s RMS models, models hosted in the Nasdaq Risk Modelling for Catastrophes ecosystem powered by Oasis Loss Modelling Framework (LMF) and many other custom and in-house models.
With the Risk Modeler application designed for catastrophe modelers, Moody’s RMS showcased two distinct approaches to integrate custom and third-party models onto the Intelligent Risk Platform:
1. Open Modeling Engine is a new Moody’s RMS modeling engine that can communicate to and operate external modeling engines from other third parties or customers. The Open Modeling Engine is used to connect Nasdaq’s Risk Modelling for Catastrophes service based on the Oasis LMF with the Intelligent Risk Platform.
With the Open Modeling Engine, customers can plug their existing Nasdaq modeling engine instance directly into the IRP. They can seamlessly execute third-party risk models available on their Nasdaq environment using the Risk Modeler application on the IRP.
2. Native Modeling Engine is another new Moody’s RMS modeling engine that can integrate custom or third-party models to be natively executed for use within the Intelligent Risk Platform.
Models using the Native Modeling Engine can use parts of Moody’s RMS native modeling capabilities and Moody’s RMS standard data formats such as the Risk Data Open Standard™ (RDOS), to natively manage data and execute models just like other Moody’s RMS models.
Similar to the Open Modeling Engine, models running in the Native Modeling Engine can be used seamlessly in applications such as Risk Modeler running on the IRP.
Although both the Open Modeling Engine and the Native Modeling Engine are still under development, several customers are helping test the capabilities with teams from Moody’s RMS, Nasdaq, and Fathom.
The workflow demonstrated at Exceedance 2023 showed a preview of how, using the Risk Modeler application in combination with the Nasdaq modelling service, users can:
Natively import Exposure Data Module (EDM), Catastrophe Exposure Data Exchange (CEDE), MRI, or Open Exposure Data (OED) formatted exposure
Run Oasis LMF-based models such as Fathom’s U.S. flood model, combined with Moody’s RMS™ U.S. Inland Flood HD Model.
Group results using the Moody’s RMS loss grouping engine to capture a holistic sense of the risks facing a U.S.-based risk portfolio.
The combined execution of the Moody’s RMS and Fathom models was performed using the Open Modeling Engine in the Moody’s RMS Intelligent Risk Platform. Users will be able to connect these services by plugging their Nasdaq platform credentials into the Intelligent Risk Platform.
This service-to-service connection allows the Risk Modeler application to execute models available from local Moody’s RMS modeling engines or through the Open Modeling Engine that connect to remote modeling services.
The Moody’s RMS Open Modeling Engine is built to perform the necessary data conversions, to bring all exposure data from the Intelligent Risk Platform to the Nasdaq modeling engine in the OED format and feeding the necessary model profile settings.
Once the Nasdaq modelling service executes the model, the Open Modeling Engine collects the Oasis Results Data (ORD)-based model output and brings that into the Intelligent Risk Platform data store built on the RDOS.
At Exceedance, we also showcased our Native Modeling Engine with sample models that allowed embedded execution of risk models using the Moody’s RMS native data format - the Risk Data Open Standard - and without the overhead of data conversions and service-to-service communications.
In one demonstration, the Risk Modeler application and the Native Modeling Engine were used to execute Fathom’s U.S. flood model as well as another example.
The unified modeling architecture is not only built to be available on the Risk Modeler application, but in the future it is expected to be accessible using IRP applications such as UnderwriteIQ™ and TreatyIQ™, two applications that execute models for underwriting use cases for primary and treaty underwriters.
With these new modeling engines that Moody's RMS are introducing, our goal is to:
Allow multiple views of risk by extending the breadth of the IRP modeling portfolio to over 400 Moody’s RMS models, more than 300 models available in the Nasdaq platform and a multitude of custom models, with a total of over 700 models available through the Intelligent Risk Platform
Seamlessly run rich collections of models using standard exposure formats such as EDM, CEDE, or OED without having to deal with the complexities of data conversions
Use Moody’s RMS intuitive financial processing for combining multiple views of risk from various models, as well as opening existing Moody’s RMS risk services such as marginal impact, exposure and loss roll-up, and many accumulation categories, scenario analyses, and so on
Simplify risk transfer with Results Data Modules (RDM), period-loss table (PLT), and year-loss table (YLT) formatted modeled portfolio result-sharing capabilities
We are excited about the way the Open Modeling Engine preserves the independence of model vendors, and for users to also bring their model natively to the Intelligent Risk Platform through Native Modeling Engine - providing a host of flexible options to our partners and customers.
These new capabilities will not only simplify a great deal of infrastructure complexity for many customers but will enable many more to have access to multiple views of risk, which I expect will also accelerate industry innovation in the risk analytics space.
For more information about this announcement and to participate in the preview program, customers can reach out to their Moody’s RMS representatives or email us at firstname.lastname@example.org.
Cihan Biyikoglu is the Executive Vice President, Product for RMS, responsible for product management across the full suite of RMS models and risk management tools. He has extensive experience in leading product management for innovative machine learning and big data analytics solutions at Fortune 500 companies over the last 20 years.
As a former Vice President of Product at Databricks and Redis Labs, Cihan both developed the product strategy and road map for open-source technologies such as Apache Spark and Redis and respective enterprise offerings in the public and private cloud platforms.
Cihan also worked on products at Microsoft, Couchbase, and Twitter, where he focused on on-premises and cloud offerings in the data and analytics space. At Microsoft, Cihan focused on the incubation of the Azure Cloud Platform in its early days and the SQL Server product line, both of which have grown to multi-billion-dollar businesses for Microsoft.
Cihan holds a number of patents in the data management and analytics space, and he has a master’s degree in database systems and a bachelor’s degree in computer engineering.