In recent years, there has been a great increase in the use of simulation techniques in fields touched by actuaries. Financial firms and others who advise or regulate them often need to estimate the value of portfolios of assets and/or liabilities. At least one side of the balance sheet is often too complicated to price analytically. In such circumstances, regulatory regimes such as Solvency II favour simulation-based methodologies to identify fair or market-consistent values. These same organisations are also typically obliged to develop risk measures quantifying the (financial) risks to which such portfolios are exposed. For larger organisations or ones with more complicated asset or liability books, this can often introduce a need for nested simulations involving so many calculations that runtimes can become important constraints on how organisations can tackle such tasks.

These challenges are often in part subcontracted to specialist teams or external suppliers. However, this does not absolve those within the organisation guiding the relevant activities from understanding what is going on, how robust are the relevant calculations and whether it is possible to circumvent runtime and other constraints in ways that may add value and reduce the costs of carrying out such exercises.

During this web session we will explore some techniques that may assist with these goals. We will pay particular attention to techniques relevant to the calculation of risk measures and other typically computationally more challenging problems actuaries can face in such exercises. The techniques will range from the conceptually simple, such as merely selecting fewer or a more representative sample of simulations or parallelising computations more efficiently, to the conceptually more complicated, such using algorithmic differentiation, segmented Monte Carlo, artificial intelligence techniques and novel computing technologies such as quantum computing. The session will set these techniques within the context of relevant regulatory guidance on how such exercises might best be carried out.

Participants are expected to be reasonably familiar with the sorts of simulation-based exercises that actuaries (and other financial practitioners) can get involved with for reserving or other purposes, and with the financial and statistical underpins of such exercises. They are expected to have some familiarity with typical programming language constructs, but it is not essential to be familiar with any particular programming language or modelling package.

Technical Requirements
Please check with your IT department if your firewall and computer settings support web session participation (the programme Zoom will be used for this online training). Please also make sure to join the web session with a stable internet connection.

The purpose of this web session is to discuss various techniques that actuaries and others needing to carry out or oversee simulation exercises can use or propose to use to improve the runtime characteristics and likely robustness of these exercises. We will interactively explore some of the techniques involved with worked examples and with reference to regulatory and other texts.

The language of the web session will be English.

Malcolm Kemp
Malcolm Kemp is a Fellow of the Institute and Faculty of Actuaries, a Chartered Enterprise Risk Actuary, Managing Director of Nematrian and a Visiting Lecturer at Imperial College Business School, London where he teaches a course in Enterprise Risk Management. He holds a first-class honours degree in Mathematics from Cambridge University. He has written three books on quantitative finance.

From 1996 to 2009, Malcolm was an Executive Director and Head of Quantitative Research at Threadneedle Asset Management. This role included responsibility for Threadneedle’s derivatives, risk management, performance measurement, liability driven investment and other quantitative investment activities. Malcolm was also chief actuary of Threadneedle Pensions Limited until 2021. From 2017 to 2021 Malcolm was an Associate in Barnett Waddingham’s life insurance consulting practice. From 2015 to 2023 he was a member of the Advisory Scientific Committee of the European Systemic Risk Board. Prior to working at Threadneedle, Malcolm was a partner at Bacon & Woodrow in their investment consultancy practice.