This is the program for the 2010 Joint Statistical Meetings in Vancouver, British Columbia.
|CE_29T||Wed, 8/4/2010, 10:00 AM - 11:45 AM||CC-2&3 (East)|
|Modeling Loss Distributions Using SAS/ETS Software — Continuing Education CTW|
|ASA , SAS|
|Instructor(s): Mahesh Joshi, SAS Institute|
|The insurance and banking businesses are always under great pressure to manage risks due to unexpected loss events in order to comply with regulations as well as to maintain a solvent and profitable business. A model-based approach of modeling risk consists of modeling the distributions of the frequency and severity (magnitude) of losses using the past loss data. This workshop focuses on modeling severity distributions, which are parametric families of continuous probability distributions with positive support. Topics include an introduction to the general problem of loss modeling, methods of estimating parameters of a severity distribution, and various statistics of fit used for evaluating model quality. Since any practical application of loss modeling accounts for censoring and truncation of the response as well as the effects of any exogenous factors, methods of modeling censoring, truncation, and regression effects will also be presented. Examples are illustrated with the new SEVERITY procedure in SAS/ETS 9.22. Attendees should have solid knowledge of probability distributions.|