Web Session "An Introduction to Economic Scenario Generators and their Validation" on 14/15 March 2022
The Economic Scenario Generators are at the core of stochastic models used by insurance companies. The applications of stochastic models are very diverse and include such applications as economic capital under Solvency II, ALM projections, dynamic hedging etc. All these applications impose different requirements upon the generation and the validation of economic scenarios.
In the web session, we begin by discussing the principles of risk-neutral modelling, where we are going to focus on equity modelling and interest rate modelling. We proceed by discussing real-world capital market modelling. Finally, we talk about ESG validation aspects relevant for Solvency 2 work and other applications.
Web Session "Practical Machine Learning Applications in Finance and Insurance" on 16 March 2022
The objective of this web session is that participants should become familiar with machine learning techniques used to solve practical problems in finance, banking and insurance. To achieve this we begin from the scratch and introduce machine learning workflows and techniques step by step: To start with, we give an overview of this interesting field with the primary focus on several techniques such as neural networks, among others. The key for an efficient application is the way of training machine learning algorithms and thus we focus our attention on this optimization as well. We strengthen our learned knowledge by focusing on several case studies: We consider an example within the Solvency II context such as implementing an internal model to calculate the Solvency Capital Requirement (SCR), but also applications to financial market such as option pricing by Monte Carlo methods or trading strategies. During our complete web session we learn how the introduced algorithms can be implemented so that the participants are able to build up their own use cases in Python at the end.
Web Session "Mathematical Modelling for Actuaries" on 6/7 April 2022
Actuaries are very experienced in modelling financial risks either stemming from population dynamics or from random events. Probability theory and statistics is their daily bread. But there are many other phenomena out in the world without having a direct financial impact but should be understood by actuaries as well. This web session is about models which typically are not covered in full by actuarial exams, but which could bring better insights to risks actuaries have to price. We will show very general approaches to set up models with applications from many different areas, whether it is medicine, construction, meteorology, biology or others. Of course, this web session can only be seen as an introduction into modelling and cannot cover all interesting models, but it should enable participants to find more in literature or develop their own ideas.
The purpose of the online training is to open the mind for problems which are by nature not actuarial but are very much linked to typical actuarial questions. It should enable actuaries and risk managers to think out of the box and find new ways to solve their challenges.
This online training is very interactive, participants are required to participate in several breakout sessions.
Web Session "Understanding the Performance of an Insurance Company: An Introduction" on 2-5 May 2022
Due to the inversion of the production cycle, the insurance business is very different from other traditional industries. Understanding, measuring and managing the performance of insurance companies is difficult due to the specific risks insurance companies must cover. It is therefore essential that you, as employee of the insurance sector, understand how your company is functioning, how its activity is measured via the balance sheet and the P&L, what are the main regulations influencing this measure, which indicators are used to assess the performance and what levers can improve this performance.
The aim of this workshop is to
- Present the functioning of an insurance company and the insurance and financial products it manages
- Explain how to read and understand the different elements of an insurance balance sheet and P&L
- Compute performance indicators used in different regulatory frameworks (Local GAAP, Solvency 2)
- Understand the impact of pricing & portfolio management, risk mitigation (reinsurance) and ALM on the performance
We will cover life as well as non-life insurance. Health insurance will not be specifically covered.
Web Session "Understanding IFRS 17" on 19/20 May 2022
The goal of the two-day web session is to provide participants with a comprehensive introduction to the new measurement, presentation and disclosure guidance for insurance contracts. It will cover life, health and non-life business, including the special guidance on direct participating contracts and shorter term non-life contracts and give useful examples.
In the web session, we will first shed a light on the context of accounting for insurance contracts within the IFRS 17 framework. We will present and discuss the general concepts behind the new model and refer to the application of valuation models like the Variable Fee Approach (VFA) and the Premium Allocation Approach (PAA). The web session will proceed with a discussion of topics specific to individual lines of business (highlighting topics still under discussion) and summarize potential approaches and solutions. We will close with an overview of methodical hot topics relevant for technical implementation seen in various European markets, share emerging market views and discuss these with the participants.
Overall, the goal is to enable participants to understand the standard and help transferring the requirements into your specific situation. It is thus intended to prepare participants for model development, implementation, testing, reviewing and consulting with management, accounting and auditors.
Web Session "Macro-Level Actuarial Reserving Models" on 20/21 October 2022
Over time, the understanding of all the assumptions behind the typically used reserving models can have grown a bit stale, and more recent developments might not have percolated all the way to the day-to-day practice. This web session will help the participants to overcome this.
The most widely used deterministic macro-level models, such as the Chain Ladder and the Bornhuetter-Ferguson model, will be discussed in full detail during this web session, but also stochastic macro-level models, such as the OverDispersed Poisson model or ODP model for example, will be covered. This entails a proper freshing up of the underlying assumptions and how the model is estimated, but also checks on determining if the chosen model is appropriate for the data at hand, and under which circumstances one should go for one type of macro-level model or the other.
In this web training a detailed overview is provided of the aforementioned models and during the practical sessions, R code is provided on how to implement most of the discussed topics, hereby rendering the participants completely autonomous after the webinar.
During the web session however, even if practical sessions will be organized, the main focus will be on the theory.
Web Session "Deep Learning with a Focus on Text Analysis" on 24/25 November 2022
In computational science, deep learning probably is one of the most heralded techniques of present time and recent history, mainly due to its versatility and impressive achievements likewise. Indeed, applications of deep learning range from beating the (human) world champion of the highly complicated Go game to the promise of deploying self-driving cars in the near future, on a large scale and all over the world.
Deep learning (DL) pertains to the field of artificial intelligence and is great at extracting and mastering the often highly non-linear patterns of a given process, whatever this process might be. The only main requirement is the availability of a large amount of data that describes the behaviour of the process under different conditions and a truckload of computational power. However, since the price of data storage and the effort of sampling data has dropped dramatically over the last years, and since Moore’s law on the increase of computational power does even nowadays not show any signs of a slowdown, fitting deep learning models that are able to produce extremely useful predictions are a reality and this already for some years now.
In other words, the time is high to also deploy this amazing technology in the insurance industry! However, the methodological framework that underlies this amazing technology is somewhat different from the statistical one that we’ve all grown accustomed to (mainly through our general love for GLM models), and the computational horsepower, needed to merely fit these models, is of an order of magnitude higher than the one needed to fit the classical statistical models.