Description: Reserving and ratemaking calculations are vital to how insurance companies perform financially. Current-day practices for predicting future loss ratio or pure premiums for reserving and ratemaking rates are not based on low prediction errors—mean squared error (MSE) criterion. This report provides mathematically optimal weights that lead to minimum MSE prediction errors, and the formulas are validated using retro-testing with real-world datasets. In this report, two main theorems are discussed, each addressing prediction of future loss ratio or pure premium and improving upon current practices for how rates and reserves are calculated.

Description: Presenters will address optimal new business written premium mix at the firm and state levels under varying competitive conditions. They will explore the use of mean-variance optimization to determine a company's optimal new business written premium mix for a given geographical area, predicting statewide product mixes using varying rate regulation scenarios, involving profit caps and measuring/testing a market's competitiveness for a given line of business. The scope of the ideas to be presented is international, applying to property, liability, individual and group life; individual and group health; disability and accidental insurance. Examples will be presented for general insurance in the United States

Description: To evaluate the risk associated with personal auto coverage, insurers have traditionally relied on information about the driver (such as age, gender and marital status), location (such as territory or ZIP code), the intended use of the vehicle (such as business or pleasure) and information about the vehicle when new (such as original cost). Insurers then added detail to driver information using sophisticated credit scoring and demographic models, and added detail to vehicle use information with telematics and usage-based insurance pricing. Now, insurers are adding detail about the vehicle itself using vehicle history. The presenters will discuss how carriers use vehicle history to more accurately price and underwrite risks. They will discuss obvious data elements, such as salvage title brands, verified average annual mileage and severe damage, as well as data that may not be as obvious but has proven predictive of loss in the past. Finally, using vehicle history as an entry point for UBI programs will be covered.