I recently worked on a challenging actuarial modeling project for a large insurance company. Actuarial modeling applies mathematics and statistics to assess risk in the insurance and finance industries (and in some other industries as well). Education and experience both help give actuaries professional qualification, and passage of professional exams is required in many countries.
Actuarial modeling includes a number of interrelated subjects, including probability, mathematics, statistics, finance, economics, financial economics, and computer programming. Historically, actuarial modeling used deterministic models in the construction of tables and premiums. The science has gone through revolutionary changes during the last 30 years due to the proliferation of high-speed computers and the union of stochastic actuarial models with modern financial theory (see https://en.wikipedia.org/wiki/Actuarial_science
Building actuarial models for clients, including internal clients, is a difficult and complex task. Much has been written on how to approach the challenge of managing a modeling project, and almost all of this guidance describes, more or less, a Waterfall model. They may use different terms, but it is the same: specification, design, development, testing, and implementation.
It's no surprise that actuarial modeling projects that follow this method suffer from similar issues as software development projects that are managed this way. My recent project was no exception. Therefore, we decided to implement Scrum for managing the modeling project.
I was charged with helping to make the transition. During the transition and after the first few months of running it with Scrum, I learned many new things. Although actuarial modeling has some similarities to software development, it is not quite the same. As a result, I adopted the principles of the Agile Manifesto
to actuarial modeling:
14 Principles of Agile Actuarial Modeling
- Our highest priority is to satisfy the customer through early and continuous delivery of a working actuarial model.
- Welcome changing requirements, even late in development. Agile processes harness change for the customer's competitive advantage.
- Deliver a working actuarial model frequently, from a couple of weeks to a couple of months, with a preference to the shorter time scale.
- Actuaries, developers, data specialists, and business analysts must work together daily throughout the project.
- Build projects around motivated individuals. Give them the environment and support they need, and trust them to get the job done.
- The most efficient and effective method of conveying information to and within a modeling team is face-to-face conversation.
- Without the necessary input data available, the model is useless.
- A working model that can be evaluated with real data is the primary measure of progress.
- Agile processes promote sustainable development. The sponsors, modelers, developers, and actuaries should be able to maintain a constant pace indefinitely.
- Continuous attention to the best model practices and good design enhances agility.
- Simplicity -- the art of maximizing the amount of work not done -- is essential.
- Models are self-documenting. Everything needed to understand the model is in the model. This allows models to be shared, understood, changed, adapted, and used.
- The best models emerge from self-organizing teams.
- At regular intervals, the team reflects on how to become more effective, then tunes and adjusts its behavior accordingly.
This was my first Scrum implementation outside of a software development or a business intelligence project, and I could not be happier with the outcome.