One of the advantages of using analytics for P&C insurance is the ability to more effectively create a rating structure or to better price premiums for a given policy. But challenges remain on how to build the best tools in this area, particularly in the development of multivariate analysis (MVA) tools.

In other disciplines, such as marketing and credit card risk, the use of predictive analytics and associated MVA tools has become common. But implementation challenges exist within the P&C industry, despite the fact that actuaries all have strong mathematical backgrounds. Why the disconnect?

The answer to this dilemma is not the mathematics being employed, but rather a lack of knowledge in using the data in the right manner to take full advantage of these techniques. In order to fully leverage the results of any MVA tool, hundreds of thousands of individual policy records with several hundred variables per policy need to be created, which is the core skill set of the data miner.

Since data mining is relatively new, not all actuaries have been trained appropriately in this area. Certainly the academic training that actuaries receive regarding mathematics and statistics is applicable for data mining. However, the most important component of this field—and arguably the one that is very resource-intensive—is the data environment. Unless one can create the right data environment for this kind of analysis, the use of MVA-type techniques is not going to deliver superior results over current pricing techniques.

Recognizing that there is a potential knowledge gap amongst actuaries within this area, many leading-edge organizations are realizing that success can only be achieved if there is a strong collaboration between data mining practitioners and actuaries.

Data mining practitioners will never have the domain knowledge and expertise of actuaries within the insurance risk area. This is particularly important if insurance rates need to be filed according to government regulations. The actuary's mathematical and insurance-sector knowledge provides the credibility that allows a certain rating structure that is acceptable to the regulators.

However, the use of MVA tools, which are the traditional domain of data mining practitioners, allows organizations to further improve their rating structures. MVA techniques are not new to actuaries, but how one creates the necessary data environment is not within the realm of their expertise.

By relying on the expertise of data mining practitioners, MVA solutions can be produced that more fully leverage this data environment. At the same time, these solutions can be tempered or modified based on the actuary's knowledge of what is acceptable to the regulators given the current insurance climate.

In a sense, the final pricing or rating solutions require a team-based approach between data miners and actuaries. This implies that data mining practitioners better understand the filing mechanisms of actuaries, while actuaries must have a better appreciation of the data environment and how it is critical in delivering an optimal MVA solution.

Special Reports /

Why Actuaries and Data Miners Should Huddle Up

Related Stories

Recommended For You