Predictive Analytics Case Study: Christine Hofbeck, FSA, MAAA
For more than two decades, Christine Hofbeck has blazed trails in the industries for which actuaries are most known-life and general insurance. While she had a conventional start to her career, a few unexpected twists and turns led her to a passion for what would become a theme throughout her career-finding problems and fixing them.
Most recently, she built a predictive analytics team at one of the world's largest financial services institutions to support the pricing team for a $5 billion portfolio covering life, disability, critical illness and more. The team uses new technology, modeling methods and sources of data to better match the right risks to the right price.
Armed with a degree in mathematics from the University of Pennsylvania, Christine spent more than a decade as an actuarial consultant specializing in retirement. In that role, Christine got a taste for pushing boundaries when she discovered an issue that led to building a new defined benefit plan for a client. The job required the actuarial work Christine knew well, but it also gave her the opportunity to get hands-on experience in other aspects of the project.
Christine could have chalked the whole thing up as a diversion from the actuary duties she loved and gotten back to her "day job." Instead, she says she learned "a beautiful, simple lesson that actuaries need not be pigeonholed into a typical role."
When a large global insurance agency hired her to build out actuarial reporting, Christine uncovered a modeling error that led to her unexpected segue into the world of predictive analytics.
The company was implementing a new underwriting risk selection predictive model, but the underwriters refused to use it because they were sure the results were skewed. The company faced huge losses if the model failed, so "I volunteered to fix it," she says.
"I had no predictive modeling background whatsoever-I didn't even know what a predictive model was. But my consulting experience taught me how to learn anything fast, and with supplemental training from expert consultants in the P&C modeling space, I quickly got to work."
Thanks to a crash course in predictive analytics and cross-departmental collaboration, she discovered the underwriters were correct. She fixed the assumption error, recalibrated the model and worked with teams across the company to implement the new solution.
The project went so well that Christine looked for other opportunities to supplement traditional methods with predictive analytics. For example, she built a comprehensive analytics practice for her company's P&C consumer lines, which led to a new role leading a team of actuaries, statisticians and data programmers implementing predictive modeling for the global casualty portfolio.
She soon noticed she was routinely solving problems that aren't typically considered actuarial in nature, but they all had a traditionally actuarial effect-helping price risks better. These new models led to more accurate outcomes benefiting both the insurer and the customers.
"I want to do everything I can to price my risks accurately," Christine says. "If I can use nontraditional data sources and techniques to better define any or all of the above, then we'll continue to keep the better risks and thank the competition for taking the rest."
Looking at her own experiences, Christine sees opportunities for actuaries, especially leaders, to push boundaries like she did. She notes the life space in particular needs leaders who understand the modeling process, can build capabilities and lead teams, even if they don't actually build models.
Actuaries, with their expertise in analytics and the predictive nature of their profession, have a unique understanding of both ends of predictive modelling. "Actuaries shouldn't just be looking for a bigger piece of pie," she says. "We should be creating a bigger pie."