Highlights from the ISEG Pricing Game 2023
Addactis, in collaboration with ISEG – Lisbon School of Economics & Management organized a Pricing Game on October 20th, 2023, for students enrolled in the Master’s program in Actuarial Sciences.
We would like to express our gratitude to all those who participated and contributed to making it possible!
The event proved to be immensely valuable for the students as it provided them with a comprehensive insight into the non-life pricing and a deeper comprehension of the ongoing challenges within the insurance industry.
We extend our special thanks to ISEG and Professor Rodrigue Kazzi, who recognized the potential of our Pricing Game as a practical application of the theory learned in the Actuarial Ratemaking course.
Congratulations and well-done to the winners Lucas Ivaldi and Cheng Chen! We hope that you have deepened your knowledge through this game, and we wish you good luck in your actuarial career!
Let’s explore the highlights of the event:
How the Pricing Game aligns Theory with Practice
A pricing game is a competition between students who must solve an actuarial problem by building a pricing project from start to finish to find the best solution.
The competition is the best way to align the theory and everything students have learned during their years of study with a real case study with a concrete example of an actuarial problem.
The main objectives of this event are, firstly, to give a global idea of the non-life pricing process and secondly to show the essential role of the actuary in risk analysis and its strong impact on decision-making. Plus, the actuarial students will be able to link modelling theory with the construction of technical premiums.
In this edition, ISEG students explored the end-to-end pricing process for a liability motor insurance business, from data preparation to rating strategy implementation.
They also understood the critical actuary’s role to point out price dislocation issues and help decide the most efficient pricing strategy for an existing portfolio. Along the game, they applied the theories on GLMs, Risk Mutualization and other insurance concepts within a realistic market set up.
Actuarial Problem and key learnings
The actuarial problem was about correcting price dislocation in a motor insurance business. The Pricing Game was two-fold: Provide a set of technical prices considering unusual risk behavior, and second, design a commercial strategy to drive profitability and business retention performance.
Crucial Insights from the ISEG Pricing Game:
- The importance of understanding heterogenous data and being able to explain systematic trends.
- The weight of profitability analysis in insurance pricing decision-making.
- How to embed technical prices within a comprehensive commercial strategy.
Winners’ Feedback:
Correlations: “We started by looking at the correlations between the covariates to have an overall view on how the different covariates were correlated according to the frequency and the severity.”
Selection of covariates: “We decided to exclude the written premium and the earned premium from the modelling. Then we also decided to exclude the year 2020 from the frequency model as there was the covid and so a very low frequency was observed in 2020. For the model of the severity, we decided to keep the year 2020 because, even if the frequency was very low this year, the severity was still in the same range as the year 2021. Then we use the pricing software of Addactis to generate a first model for the frequency and the severity. According to the test statistic implemented in the software we excluded some variables from the model that were not statistically significant, and we aggregate the classes of categorical variables that were not statistically significant.”
Trend: “For the trend we assume that the trend between the years 2022 and 2023 would be the same as the one between the years 2020 and 2021.”
Inflation: “Finally for the inflation, we searched on the Internet the actual inflation rate in Saudi Arabia for the year 2023.”
Commercial strategy: “For the commercial strategy, we looked at the impact of the different covariates and the different categories of the categorical variables on the premium to test lots of combinations of penalties and rewards. We started with the covariate that had huge differences between the different categories (it was the Region). And then we fixed rewards and penalties and then we tested for other covariates and so on.
Finally, we decided to keep a combination with a retention higher than 80% and not the one with a bigger profit and closer to a retention of 80%. We made that choice to avoid the randomness of the real portfolio and to be sure to have a retention over 80%.”
Lucas Ivaldi and Cheng Chen, actuarial students at ISEG Lisbon
About addactis® Pricing Games
Addactis organises “Pricing Games”, a competition between students, academics, or practitioners who must solve a real-life actuarial problem using statistical modeling techniques to estimate pure premiums under different contexts.
Find out how Addactis organizes business pricing game by reading our latest article.
Suggested Content
Don’t stop learning! Check below Addactis latest resources on Pricing:
Top 6 myths of non-life pricing insurance
Addactis experts have selected 6 Myths around pricing, which seem strangely persistent and widespread within all types of non-life insurance companies. Access now our E-book to find out those myths and how to overcome Pricing challenges in an insurance company.
Inflation: what pricing challenges for 2023?
The strong inflationary drift, together with the climatic drift, are thus pushing insurers to change their tariff structure. Read our article and find out all you need to know about Inflation, Pricing Challenges and Addactis experts’ expectations for 2023.
Addactis Pricing Game key learnings on The Actuary September Issue
The Actuary shares the key lessons from addactis last motor insurance Pricing Game, which is a competition in which actuarial professionals work together to solve a problem that reflects actual market conditions.