Unifying Modeling Architecture to Operationalize Multivendor Modeling Workflows with Moody’s and Nasdaq
Cihan Biyikoglu
James LayJuly 23, 2024
In today's increasingly complex and interconnected risk landscape, firms seek efficiencies in how their underwriters, catastrophe modelers, and exposure managers price, select, and manage risk.
Risk models provide great insights across all these critical decisions. However, to access multiple views of risk from various model vendors, firms have to develop integrations across multiple modeling systems, multiple exposure and loss schemas.
Firms experience implicit and explicit costs when running multivendor model environments, such as completing required training on the new tools used to reconcile vendor exposure and results schemas.
These schemas include Catastrophe Exposure Data Exchange (CEDE), Exposure Data Module (EDM), Open Exposure Data (OED), Event Loss Tables (ELT), Period Loss Tables (PLT), and Year Loss Tables (YLT).
The time and cost to do this reconciliation can then drive up expense ratios, and eat into the productivity of the entire risk organization. Combining Moody’s Intelligent Risk Platform with the Nasdaq Risk Modeling platform will greatly simplify this complexity.
For over 30 years, Moody’s RMS modeling has led the catastrophe risk industry, an industry we helped to pioneer. From our first California earthquake model in 1989 to today’s global suite of high-definition, climate change, and cyber models, we help insurers gain greater clarity and certainty in managing catastrophe risk with over 400 models available on the Intelligent Risk Platform.
With the release of the new Open Modeling Engine (OME), the Intelligent Risk Platform now unifies modeling under one streamlined workflow, with built-in support for all major exposure and loss result formats. This offers the scale needed to handle the weight of over 700 risk models, including 300 from Nasdaq’s Risk Modeling for Catastrophes platform.
Using the new OME within the Intelligent Risk Platform to natively integrate with Nasdaq’s Risk Modeling for Catastrophes solution, allows clients to:
Convert exposure schemas to and from CEDE, OED, and EDM so that users can run Oasis and Moody’s RMS models on the same exposure data.
Execute models within Nasdaq’s solution from Moody’s Risk Modeler™, allowing catastrophe modelers to manage and adjust the parameters of their Moody’s RMS and third-party models within a single application.
Transform results from Oasis into Results Data Models (RDM) to facilitate data exchange with broker or reinsurer partners.
Let’s look at how the Moody’s/Nasdaq partnership helps to deliver the enhanced, unified modeling workflow our clients and the market have been asking for.
Headwinds to Unifying the Multivendor Workflow
The entire insurance value chain – from specialty underwriters to the largest global reinsurance brokers – benefits from multivendor model workflows, providing access to supplementary views of risk or insights on specific perils that are complex to model.
However, adopting a multi-vendor modeling approach has historically been hindered by operational challenges Whether these are justifiable challenges or a perception from the risk management function, the unified Risk Modeler/Nasdaq approach can address many of these concerns head-on.
A recent Nasdaq working group of catastrophe risk managers and practitioners highlighted several areas where they believe challenges in adopting a multivendor workflow exist. Let’s examine these areas and how the new approach can help:
1. Challenges with Managing Multiple Platforms Limit the Ability to Adopt Models
Utilizing multiple model vendors has meant dealing separately with each vendor’s system, as no single vendor has overcome the challenge of building a unified system that could bring all model results under one consistent view. This meant firms had to create a new single-destination system to house the data required for all their modeling needs.
With all this complexity, it is no surprise that I.T. challenges often arise as an area of concern. This is addressed head-on using cloud-based systems with Moodys RMS and Nasdaq models delivered using a Software as a Service (SaaS) experience without deployment into a company’s I.T. systems.
Benefiting from Moody’s Intelligent Risk Platform integration with Nasdaq, the Risk Modeler application has the scale needed to enable access to multivendor views of risk with a streamlined modeling workflow that works for both Moody’s RMS and Nasdaq models.
2. Working With Different Exposure Data Formats
Getting data converted between different formats continues to be challenging for the industry. It is time-consuming and requires an in-depth knowledge of how the formats map to each other.
The ideal solution is to hold exposure data in a common format and push the relevant data schema into the models as required. This is easier said than done and can take considerable time and effort to accomplish, together with the ongoing maintenance as data formats evolve will also need to be addressed.
With Moody’s RMS/Nasdaq solution, the Intelligent Risk Platform will be able to use Moody’s data conversion schemas to run models on Nasdaq’s platform without the need to map data to different model formats.
This addresses the issue of maintaining data format conversion tools and enables firms to leverage a solution that has been purpose-built to deal with different data formats.
3. Model Evaluation And Validation
On top of the many other tasks that catastrophe modeling teams need to perform, evaluation and validation of models can also be time-consuming, and in many cases, lead to significant delays in adopting models.
While many of the commercial model developers provide very good model documentation, there is a need for exposure managers to understand in some detail how different models represent their risk. This entails an in-depth analysis of the modeled losses compared to their portfolio of risks.
Centralizing access to models across multiple vendors and working consistently with modeled results not only simplifies the model validation process but also accelerates model adoption, to reduce the time to value firms can achieve with an informed approach to risk selection.
4. Internal Stakeholder Buy-In
Finally, multi-vendor modeling is not just a technical challenge, but often a challenge for the people working within a firm. Educating internal stakeholders on the value of alternative or supplemental views of risk is essential to the new model adoption process.
Working with model vendors can be a big part of this process, but consistency when representing and comparing modeled outputs is essential to telling the story.
There is still a perception that including multivendor views of risk in the risk management process is only possible for the big industry players with bigger teams and more resources to utilize these models. While this may have been true in the past this doesn’t have to be the case moving forward.
Operationalizing the Multi-Vendor Model Workflow
Introducing the new Open Modeling Engine on Moody’s Intelligent Risk Platform will allow connection to Nasdaq’s Risk Modelling for Catastrophes service, enabling catastrophe modelers to communicate to and run Nasdaq-hosted models on Risk Modeler.
The Moody’s Open Modeling Engine executes the required data transformation and extraction of the loss tables between the Moody’s and Nasdaq platforms, as shown in the figure below, users utilizing Risk Modeler select from a combined list of over 700 models without the complexities mentioned above.
In addition, Moody’s RMS model and third-party model results stored in the unifiedRisk Data Lake can utilize downstream Moody's financial engines for post-processing (Grouping, Post-Analysis Treaty Editing (or Editor) (PATE), and so on) to deliver a consistent set of financial analyses.
By removing many steps in the multiple model workflow, catastrophe modelers will have much more time to validate model results, craft new views of risk, and rerun and tune their model outputs, all of which enhance the insights provided to stakeholders and improve the business decision-making process.
To learn more about third-party modeling on the Intelligent Risk Platform or to sign up for our preview program, ask your customer success representative or explore our Third-Party Modeling web pages here.
Managing Director - Head of Product for Moody's RMS
Cihan is the Managing Director - Head of Moody's RMS Product, responsible for product management across the full suite of Moody's 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 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 into multi-billion-dollar businesses for Microsoft.
Cihan holds several 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.
As the Commercial Director of Nasdaq Risk Modelling, James leads the business globally which includes commercial engagement with Nasdaq Risk Modelling clients, as well as with risk model providers that collaborate with Nasdaq via the Nasdaq platform.
He has been working in the field of catastrophe modeling technology since 2012 and has over 20 years of experience in the technology industry, including cloud and data analytics platforms.
While at Simplitium/Modex, James was involved in the Oasis Solutions Project, helping deploy the technology and partnering with a consortium of businesses to establish the first independent cat modeling platform available to model providers and model users alike.