Welcome
April 2024
Welcome to our first newsletter of 2024. It's been a busy start to the year for Oasis which is set to continue over the coming months. We always want to engage with the cat modelling community, so please get in touch with any feedback or questions.
Max Info release Magniphi, their Automated Model Validation Application
In February 2024, Max Info launched Magniphi: a web-app that connects to stochastic event set information and catastrophe model generated loss tables to create the first vendor/broker-independent, multi-peril cat model evaluation platform.
- Magniphi produces auto-generated reports to remove the manual effort required to produce the exhibits and data cleaning during model evaluation processes. These reports can facilitate regulatory responses in addition to encouraging progressive and transparent dialogues with vendors and brokers.
- Magniphi is fully interactive, with cleaned and transparent scientific data, methods and information. This enables users to quickly up-skill themselves and uncover the many unknown-knowns generated in the cat modelling process.
- The peril-region coverage is currently: US hurricane, US severe convective storm, US earthquake and US wildfire. The next phase of development planned towards the end of 2024 will extend the peril scope to European windstorm, US and European flood.
- The underlying engine can also be used for climate change reporting and live-climate state model adjustments.
- Utilising the Oasis framework and Nasdaq's Risk Modelling for Catastrophes (NRMC) interface, models can be evaluated autonomously enabling easy and quick access to multiple models.
- The Magniphi app has an 'early-adopter' pricing offer, which is open until 30th June 2024.
Get in touch with the MaxInfo team at hello@maxinfo.io to find out more.
New Hail Model for France from Insurance Partners Europe (IPE) - now Available in Oasis
In 2022, France experienced significant loss to hail storms with a total market Insured Loss reaching €6bn which is the second largest loss from a nat cat event behind windstorms Lothar and Martin in 1999 that had estimated insured losses in France of around €9bn (2023 projection) (https://www.franceassureurs.fr/).
Some key considerations for hail in France are:
- The frequency of hail storms is increasing across France but especially in regions that historically have not experienced significant loss to hail, such as Brittany.
- Hail storms are highly seasonal and clustered so they can occur across several, consecutive days.
- The impact of climate change is considered as a contributing factor.
- Hail is the main cause of damage to the motor classes.
IPE has developed a hail model for France which is now available on the Oasis platform and features:
- Hazard embedded clustering that allows to price reinsure time clauses.
- The impact of climate change has been implemented, allowing short, medium and long range projections.
- The new damage functions developed in this model provide a more appropriate approach to modelling the motor class.
- The model will extend to cover Italy in Germany by the end of 2024.
Contact Pierre Mouilhade if you have any questions (pmouilhade@inspart.eu).
UCL and Actuari looking for Collaborators to Pioneer Federated Computing Infrastructure
Oasis has been working with University College London (UCL) and a spin out called, Actuari, exploring ways of how new technology can help solve problems. One in particular is Federated Learning which works by training a central model across decentralised devices or servers. Instead of moving all data to a central location, the model is trained locally on each device, and only the model updates are shared. This maintains privacy and allows collaborative learning without sharing raw data.
UCL and Actuari invite forward-thinking insurance companies to join us in pioneering Federated Computing Infrastructure that will shape the future of insurance analytics. We welcome your collaboration in exploring innovative use cases where privacy-preserving technology infrastructure can bring value to insurance. As early adopters, you'll have the opportunity to shape the future of insurance analytics alongside us.
Key Features of Federated Computing Data Infrastructure:
- Privacy at Its Core: Our cutting-edge infrastructure prioritizes privacy, enabling secure collaboration without compromising raw data integrity.
- Bringing Algorithms to Data: Through Federated Machine Learning, we bring algorithms to the data, ensuring privacy and security while empowering companies to leverage collective intelligence.
- Proven Results: Extensive research validates secure collaboration on ML models without the need to move or share raw data. Actuari is transforming this research into a robust software solution.
Join us in pioneering the future of insurance analytics. Contact us today at margo@actuari.eu
Moody's RMS Announce the Integration with Nasdaq Risk Modelling for Catastrophes (NRMC) will go live on 30th June
As mentioned in our December newsletter, we fully support the integration of Oasis-based models through the RMS Intelligent Risk Platform (IRP) via Nasdaq's Risk Modelling for Catastrophes (NRMC) interface. This opens up the market for users of all models.
Full article can be found here:
https://www.rms.com/blog/2024/03/01/introducing-unified-modeling-on-the-intelligent-risk-platform-with-moodys-rms-and-nasdaqs-risk-modelling-services
Get in touch if you have any questions or enquiries.
|
|
Oasis Insight Conference
Bermuda
12th June 2024
Royal Bermuda Yacht Club, Bermuda
Registration will open in due course but a draft agenda can be found here:
https://oasislmf.org/events/oasis-insight-bermuda