Skip to content

stufraser1/OpenDataStandards

 
 

Repository files navigation

Latest version of ODS ODS version

                                                       

Mission Statement:

ODS should be the standard used for exposure data, result outputs and contract(s) terminology for the (Re)Insurance industry and other interested parties in catastrophe modelling and exposure management across all classes of business.

Steering Committee Members:

                                                                                                       

 

Open Data Standards (ODS)

ODS is curated by Oasis LMF and governed by the Open Data Standards Steering Committee (SC), comprised of industry experts representing (re)insurers, brokers, service providers and catastrophe model vendors. The SC will evolve over time and include Subject Matter Experts to assist with areas requiring specific domain experience, including data validation and support for enhancing and automating the interoperability of these standards.

The components of ODS are the Open Exposure Data (OED) format and the Open Results Data (ORD) format. Both OED and ORD are designed to assist with solving interoperability problems current in the insurance market, where implementing a model-developer-independent exposure data and results repositories will assist in creating choice in the use of catastrophe models and analytical tools.

Further information and community views of ODS can be found on the ODS website: https://oasislmf.org/open-data-standards

 

Structure of ODS

The diagram below highlights the proposed, long-term structure of ODS and all the key components. Interoperability is vital to ensure efficient interaction across multiple databases, systems and external exposure management and data storage facilities.

 

 

ODS Implementation into Oasis: Oasis LMF are continuously expanding the ODS functionality they support on their platform, especially in their financial module (FM). The tracking of this work can be found here: https://github.com/orgs/OasisLMF/projects/5.

 

Open Exposure Data (OED)

The aim of OED is to provide the industry with a robust, open and transparent data format. This will improve efficiency and transparency for the cat modelling community, facilitating data transfer and analytics across models and vendors. OED is a model agnostic data format that comprises of four input files (Acc, Loc, RI Scope, RI Info).

Detailed descriptions of each data field are covered in the Open Exposure Data Spec.xlsx and each tab is split out into .csv format and located in the 'schema' folder.

Detailed reference and background OED information can be found the 'docs’ folder. Examples of how to code multiple financial structures in the input files are covered within these documents.

The web (HTML) version of the OED documentation can be viewed here:

https://oasislmf.github.io/OpenDataStandards/index.html

OED is the exposure data input format supported by the Oasis Loss Modelling Framework and by all models deployed on the Oasis platform.

 

Open Results Data (ORD)

ORD was initially developed during the Lloyd's Lab innovation project (Cohort 3) in 2019, by a working group led by Oasis, that focussed on constructing model agnostic results formats and appropriate data formats. These model outputs will cover an extensive suite of results that can be isolated by aspects of the exposure data, financial and statistical perspectives. The ORD Standard encompasses:

• Exposure data

• Exposure Summaries

• Modelled Losses

• Loss curves

• Other metrics

• Analysis meta-data

Working documents can be found in the 'schema' and 'docs' folder (https://github.com/OasisLMF/OpenDataStandards/tree/develop/OpenResultsData)

ORD will be implemented by the Oasis Loss Modelling Framework during 2020/21 and all updates and progress will be located in this repo.

 

Governance

ODS is governed by a steering committee that meets periodically and is chaired by Oasis LMF. The main objectives of the SC are to:

• Agree Governance, Membership and Terms of Reference of SC

• Oversee and provide direction to the continued development of Open Results Data

• Oversee the curation of the Open Exposure Database

• Establish Working Groups, where appropriate, to monitor and manage the day-to-day detail of the technical aspects of data transformation

• Coordinate with stakeholders a pathway for the adoption of ORD

In addition, the SC determine the design of the data standards, make decisions about the technical implementation and prioritise work. Working groups may be formed from time to time to manage specific technical enhancements (e.g. the design of a new major feature).

The committee is made up of representatives from the following organisations:

Technology Providers: Nasdaq

Cat Model Vendors: AIR Worldwide, CoreLogic, Impact Forecasting, JBA Risk Management

Insurers: Ascot Underwriting, Zurich

Reinsurers: Renaissance Re, Swiss Re

Brokers: Aon, BMS, Guy Carpenter, Willis Towers Watson

Market & Data: Insurance Development Forum (IDF), Lloyd's of London, Perils AG

A maintenance group will maintain and update the assets that define the data standards. In the first instance, Oasis LMF will provide the maintainers, though this could be extended to include members of the user community.

 

Releases

All the assets that define ODS will be managed in this GitHub repository. Minor releases will occur on a quarterly basis and will be backward compatible, and major releases will occur on an annual basis and require a change management process. The releases will follow the SemVer convention (https://semver.org/), so given a version number MAJOR.MINOR.PATCH, increment the:

  • MAJOR version when you make incompatible changes e.g. changing column names, changing the structure of the data.

  • MINOR version when you add functionality in a backwards compatible manner e.g. adding a new column with a default value, adding a new allowed value for an existing field.

  • PATCH version when you make backwards compatible bug fixes e.g. correcting a typo in a column label.

All new work will be done in feature branches, following the GitFlow model. The latest released version will be held in the master branch and the current development work will be in develop or specific feature branches.

 

Tracking

All new work will be captured as issues in this repository. Any GitHub user can raise an issue and the maintenance group will classify as:

Major Features: These are major work items that will require significant work and may cause breaking changes. An example of a major feature would be extension to a new class of business. All major features will be reviewed during a steering committee meeting.

Minor Features: These are minor work items that will not cause breaking changes. An example of a minor feature would be the inclusion of a new location attribute with a default value. Minor features must be approved, by email, by two members of the steering committee before being addressed by the maintenance team. If approval is not given, or an objection is raised by a member of the steering committee, then the feature will be reviewed at the next a steering committee meeting.

 

Documentation Updates:

These can be actioned directly by the maintenance group.

 

Bugs:

These can be actioned directly by the maintenance group, assuming they do not cause a breaking change.

 

Attributions

OED was developed by a working group of industry practitioners in close collaboration with AIR Worldwide who generously made their CEDE schema available to the industry. The working group was chaired by Matthew Jones with technical work led by Aiste Kalinauskaite both of NASDAQ (formerly Simplitium). The initial OED assets were ported from https://github.com/simplitium/oed on 28 April 2020 with permission from NASDAQ.

 

License

The content of ODS including the documentation and the schema within both Open Exposure Data (OED) and Open Results Data (ORD) are licensed under the CC0 1.0 Universal license. Any code, MS SQL scripts or the development of tools are licensed under BSD 3-clause license.

About

Open data standards curated by Oasis.

Resources

Stars

Watchers

Forks

Packages

No packages published

Languages

  • Python 87.0%
  • Makefile 8.2%
  • Shell 4.8%