Predictive analytics is a significant advancement in the ongoing search for better ways to develop equitable rates — rates that accurately reflect the experience of groups of insureds. Some companies are building their own predictive models. That’s a major undertaking, requiring a significant commitment of resources, time, and expertise that few companies have. Others are jump-starting their modeling efforts by using a predictive model that’s already built and tested.
Whichever option you choose, you need to evaluate how a predictive model would affect your underwriting decisions. In the Verisk Review article titled “Adopting Predictive Modeling for Competitive Advantage,” Glenn Meyers, Ph.D., FCAS, MAAA, addresses some key questions you should contemplate during your evaluation:
- If you use the model to change prices, how much would prices change?
- Can the model systematically identify risks that have lower or higher loss ratios under your current rating plan?
- How do you compare the costs of implementing a predictive model with the benefits it provides?
He adds that although developing a model can be a lengthy and costly process, there can be significant long-term benefits. Implementing advanced predictive modeling tools to streamline products, prices, and services can prove worthwhile, particularly in today’s demanding and interdependent economic environment.