Accountable Algorithms: Insurance Hackathon

February 28, 2022
VIRTUAL EVENT - this is an ongoing hackathon. The event date/times stated are therefore flexible and you are able to complete the tasks at a time convenient to you, before submission on 11 March, if you register before 28 February
Enter Here

Description

Please note that this event runs throughout February and March. The event date stated is the closing day for registrations, but registered participants can partake during February and March.

Audience

The programme is suitable for anyone who:

  • would like to take part in an insurance related hackathon focussed on model risk 
  • wish to develop accountable algorithm and modelling techniques
  • want to network with, and learn from fellow participants
  • is interested to gain experience presenting their model to stakeholders  

Team or individual entries possible. Limited spaces available.

Details

Date: 9 February - 31 March 2022 

Time: Work in your own time and drop in to optional Q&A sessions with the organisers.

Hackathon submission: 11 March 2022 by 20:00 GMT via the online platform. Late submissions will not be accepted.

Shortlisted presentations: Successful teams will be required to present live to a panel on 25 March 2022. 

Platform/Venue: Register & submit your attempt online, Q&A sessions and live presentations via Zoom.

Fee: £10 per entry to cover administration costs

CPD hours: This event can count towards your IFoA CPD requirement if you consider that you have learned something relevant to your current or future actuarial work or professional development, that it addresses a personal development need and has an identifiable learning outcome. 

Registration closes 28 February 2022.


A calendar invite for optional drop-in Q&A sessions will be sent. A placeholder invite will be sent for 25th March and successful participants will be notified by email if they are required to attend and present.

Outline

This hackathon requires participants to build a model that aims to predict whether a motor insurance policyholder will renew their policy at the end of a given calendar year based on anonymised policyholder data. 

Participants will need to present their investigation as if they were advising insurance executives. They should be able to explain & communicate their approach & methodology as well as the implications of their model. This should include considerations of the model risk for the organisation; including uncertainties and implications of their model, conclusions, recommendations and findings. Reference to professional guidelines may be made.

Organisers will review submissions and select based on the criteria outlined above a shortlisted set of submissions, to present to a panel of risk management and insurance executives.

Timeline

For more information, or to enter, please visit https://training.actuartech.com/courses/actuartech-qed-hackathon



Newsletter

Keep up to date

Sign up to our newsletter for the latest in actuarial technology and data science

Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.