You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
In the context of FINOS TechSprint 2024, the goal is to develop a proof-of-value (POV) for the business and proof-of-concept (POC) for the technological feasibility that simulates the effects of physical risk models and transition risk models on market risk and credit risk in trading activities, taking as the initial example a securitised portfolio of mortgages (MBS). More information here finos/traderX#191
Mortgage Back Securities (MBS) are estimated to be around $10 to $12 trillion in total value. This data is relevant for banks, insurance companies and asset managers. The data can be used to build climate-focused applications and AI models, such as predictive and anomaly detection models, that can assess the climate risk exposure of borrowers more accurately. By analyzing patterns in loan repayments, borrower behavior, and climate-related factors (such as extreme weather events or carbon footprint), these models can identify early signs of financial distress driven by climate risks.
Other Comments
POC:
David Kelly (BIP) is TechSprint Use Case Project Lead
John Mark Walker (Fannie Mae) is FINOS AI Strategic Initiative Executive Sponsor
Others:
This issue is part of FINOS LLM Exploration Project and aims to validate the fit-for-purpose of OS-C Data Commons (and in particular Data Request) as "A comprehensive data contribution framework document that financial institutions can follow to contribute data securely and efficiently to an open financial dataset." More info here (slide 17)
The text was updated successfully, but these errors were encountered:
Thanks for creating this record of interest. To set expectations, this data request form creates a potential work item that the OS-Climate community can prioritize, but the OS-Climate project does not make any promises that it will onboard the data.
@ericbroda copying in Eric Broda, who may have some advice on such data might be federated via the Data Bazaar. It would still require somebody to take ownership of doing the federation, but the Data Bazaar metadata and management might present a more favorable ROI equation than simplying trying to onboard the data without such structural/architectural support.
What dataset are you requesting?
US residential mortgage perforamance dataset from Fannie Mae Data Dynamics.
Is this open data?
No, currently available under Royalty-free data license
What is the use case for this dataset?
In the context of FINOS TechSprint 2024, the goal is to develop a proof-of-value (POV) for the business and proof-of-concept (POC) for the technological feasibility that simulates the effects of physical risk models and transition risk models on market risk and credit risk in trading activities, taking as the initial example a securitised portfolio of mortgages (MBS). More information here finos/traderX#191
Which organization produces the dataset?
Fannie Mae
Which format is the data in?
Refer to the links above
Which format would you like the data in?
TBD
How often does the data update?
Here the link to the refresh calendar
Who else would use this data ?
Mortgage Back Securities (MBS) are estimated to be around $10 to $12 trillion in total value. This data is relevant for banks, insurance companies and asset managers. The data can be used to build climate-focused applications and AI models, such as predictive and anomaly detection models, that can assess the climate risk exposure of borrowers more accurately. By analyzing patterns in loan repayments, borrower behavior, and climate-related factors (such as extreme weather events or carbon footprint), these models can identify early signs of financial distress driven by climate risks.
Other Comments
POC:
Others:
This issue is part of FINOS LLM Exploration Project and aims to validate the fit-for-purpose of OS-C Data Commons (and in particular Data Request) as "A comprehensive data contribution framework document that financial institutions can follow to contribute data securely and efficiently to an open financial dataset." More info here (slide 17)
The text was updated successfully, but these errors were encountered: