Many P&C insurance companies are seeing deteriorating underwriting results. With the advent of comparative raters in the P&C insurance market, prospects can compare prices on many companies instantly, often simply choosing the lowest offer. Less sophisticated insurance carriers become exposed in the areas where they are mispriced to make a sale. The lowest cost may win the business, but may be underpriced relative to the risk.
Actuaries and pricing departments dedicate hours to getting more accurate models, but it’s hard to fight with other insurers if everyone uses the same standard approach. And it’s worth noting that insurance pricing does not only belong to the GLM anymore. AI in insurance is becoming more prominent. From preparing the data to real-time pricing, our platform can help actuaries move to data-driven pricing.
Ming-Li Gridel, Insurance Customer-Facing Data Scientist will demonstrate:
- How to prepare data in DataRobot Paxata from several datasets
- How to automatically build machine learning pricing models, including exposure, monotonic constraints, and frequency severity approach
- How to extract a GAM rating table or deploy a real-time pricing machine learning model
