CAS Co-Sponsored Research Addresses a New Way to Look at Personal Auto

Auto Loss Costs: Dynamic Linear Models with Changepoints

In this paper a new model is developed to better understand the complicated dynamics in personal auto insurance. A dynamic linear model with seasonality, regression on congestion, and a linear trend with a changepoint is designed. The changepoint enables modeling of structural shifts in the industry, regardless of why they occur (e.g., regulatory, economic, or social changes). This research is sponsored by the CAS, the SOA and the American Property Casualty Insurance Association.

The changepoint improves the model fit and will likely lead to improved predictions of future losses; urban congestion best describes the loss process; frequency has generally decreased; and severity has generally increased. Loss cost has mainly increased, but a large number decreased at the beginning of the time window. The changepoint allows the slope and intercept to change, improves the model fit and will likely lead to improved predictions of future losses. This model is likely better able to model the uncertainty in the industry going forward.

READ THE PAPER

For earlier research reports on auto loss cost trends, please check the third report, second report and first report.

For more information, contact Brian Fannin, CAS Research Actuary, at bfannin@casact.org.