Description: This article proposes a framework for estimating mortality improvement models and analyzing the parameter uncertainty in such models.

Description: Explores the open vs. closed systems debate with regards to actuarial modeling and exposes the over-simplification. Valuable for anyone who uses models.

Description: Presented at October 1996 Annual Meeting. An update on Futurism Section activities followed by an introduction to common techniques of futurism. Examples are given how actuaries can use these techniques in their day-to-day work.

Description: This article describes the use of artificial society modeling to gauge the effect of insurance agent population and effectiveness on individual life insurance sales. This is part of an ongoing effort to extend the use of one type of agent-based modeling beyond health care, but is far from a practical application at this point.

Description: This article describes some of the drawbacks to the assessment of longevity risk adopted by most practitioners within the UK together with an initiative that provides a platform for an alternative approach that is both (a) more forward-looking in nature and (b) can complement current practices.

Description: The article describes the key components of actuarial model, leading practices in designing them, and characteristics of well-built models.

Description: The article gives a brief history of the development of the Delphi method, and describes the use of the Delphi method in studies by the Society of Actuaries over the past 25 years.

Description: Description of the vast variety of predictive analytics topics covered by the Forecasting & Futurism section and introduction to the July 2015 issue of the newsletter.

Description: This teaching session is about conveying to users of models and forecasts how uncertain the outcome of stochastic projections really is.

Description: This 1979 session is about recent advances in Prediction Theory. Topics included: 1. Statistical methods to predict path properties of a Time Series. 2. How and to what extent Box-Jenkins Time Series methods can be modified to predict two or more time series simultaneously. 3. Workable techniques that are already being successfully used in practice.