TIGI 2020 (Technical Issues in General Insurance)

09.00-10.00: Lloyd’s Validation Review Outcomes 2019

The Lloyd’s syndicate capital team would like to present the findings of the recent market validation review.  The presentation will cover themes from across the market, such as common areas where improvements are required, as well as some examples of best practice.  We intend to look at a few areas of validation in some detail, as opposed to covering an overview of all areas. 

Speakers: Mirjam Spies and Rebecca Soraghan, Lloyds

A recording of this webinar is available for members and non-members to purchase.

Video costs: Members: £15 | Non-members: £35 

09.00-10.00: Backtesting Dependencies: Maximising the Value of your Historical Data

Backtesting is one of the more valuable tests performed when validating internal models: it tells us something about model output relative to results observed historically (hopefully that the two are in agreement!)

However, validating dependencies in this way is often challenging.  For any pair of risks, we can calculate the historical correlation, but this only provides a single data point to rely on.  There is also no guarantee that this figure will bear much resemblance to the selected correlation being modelled.

Our overall aim is to discuss a framework for validating this aspect of capital models in a useful and structured way, based on a recent case study.

The main areas of the talk will be as follows:

• A quick review of the main dependencies backtesting methods currently available (implicit and explicit)
• An extension of the explicit method, based on assessing the extent to which ranges of inputs/outputs fall within confidence intervals around historical values
• A further extension of the explicit method, in which we examine how well backtesting works for various segments of a typical insurance correlation structure

We will also discuss some potential other uses for the methodology, focusing on enhancements to a typical ESG validation exercise.

Speaker: Neil Gedalla, LCP

A recording of this webinar is available for members and non-members to purchase.

Video costs: Members: £15 | Non-members: £35 

12.00-13.00: Pricing like a Data Scientist

The presentation will cover a summary of my findings over the past few years of working on the boundary between data science and actuarial. I will describe and review several data science approaches and discuss their relevance and use in actuarial pricing. I will present a list of recommendations ranked by reward/effort. Examples presented will be based on the R statistical language but the presentation will not be code-heavy, more a discussion of approaches and techniques, and importantly, what we can learn from our colleagues in the data science field.

Approaches discussed will include:

• Exploratory Data Analysis ("EDA")
• Clustering and Dimensionality Reduction
• GAM's and GLM's
• Imputation
• Cross-validation
• Bootstrapping
• Using machine learning models even if legacy-constrained
• Automation and the cloud.

Much discussion is around GLM's and related techniques but examples from commercial lines will also be included, where the issues of data paucity and accuracy are more important.

Speaker: Matthew Evans, EMC Actuarial and Analytics

A recording of this webinar is available for members and non-members to purchase.

Video costs: Members: £15 | Non-members: £35 

11.30-12.30: Machine Learning in Reserving: Practical Applications and Lessons Learned

Machine learning can significantly enhance insurers’ reserving processes. However, moving to business as usual can be challenging.

We will openly share practical applications and lessons learned from embedding machine learning in reserving, including:

• Key opportunities for machine learning with insurance reserving.
• Examples of embedding machine learning – covering both commercial and personal lines insurers.
• Key challenges and lessons learned.

No prior technical knowledge of machine learning techniques is required.

Speakers: Tom Durkin, LCP and Hazel Beveridge, Pioneer 

A recording of this webinar is available for members and non-members to purchase.

Video costs: Members: £15 | Non-members: £35 

09.00-10.00: Best Practices for Developing and Deploying Web-Based Pricing Tools

Across the General Insurance market actuarial pricing tools are being migrated from Excel to web-based technologies. Web based pricing technology offers many synergies with AI and Machine Learning. But they are often hampered by poor design and practically always perform worse than their Excel counterparts!

The following will be presented:

• Best practices when moving from Excel pricing tools to web (eg. parameter management, data-modelling, actuarial calculation design)
• Demystifying the web technology stack by example: demonstration of key components (UI, database, calculation engine)
• How web-based pricing tools can be integrated with 3rd party platforms (eg. Lloyds risk exchange, Internet of Things etc)

This talk will be accessible without prior knowledge of web-technology or actuarial pricing.

Speaker: Edward Anderson, Probitas 1492

16.00-17.00: Lloyd’s Casualty Market Study

Lloyd’s will be presenting on the recent Casualty Market Study that was conducted across a section of the market. This study looked at the Non-Marine General Liability, Directors & Officers and Medical Malpractice from an underwriting, reserving and claims perspective. The presentation will cover the reasons for why the Study was conducted, the approach taken to determine the focus classes and participants, the key general findings as well as the next steps that Lloyd’s are proposing to take.

Speaker: Ajay Shah, Lloyds