Description: This paper addresses the fuzziness in regression models. The goal is to present a test procedure to explicitly examine whether an independent variable has a clear functional relationship with the dependent variable in a specific regression model, or whether their relationship is fuzzy. To this end, the authors interpret the spread of the regression coefficients as a statistic measuring the fuzziness of the relationship between the corresponding independent variable and the dependent variable. They then derive test distributions based on the null hypothesis that such spreads could have been obtained with data generated by a classical regression model with random errors. The analysis is presented in conceptual rather than technical terms.

Description: In session 14 MM of the Washington Annual Meeting the instructor summarizes the research topics presented at the actuarial research conference held at the University of Calgary during August 1997. From the Record of the Society of Actuaries, Vol. 23, No. 3.

Description: This paper examines four and-a-half years of prescription records for an elderly population who have relatively comprehensive out-of-hospital drug coverage under the Pharmaceutical Assistance Contract for the Elderly program in Pennsylvania. The result is a discussion of the determination of non-parametric estimations of the distributions. From Actuarial Research Clearing House 1991, Vol 1.

Description: Genetic algorithms [GAs] and neural networks [NNs] are currently being used in actuarial and financial modeling. Nonetheless, while there is a general awareness of the nature of these techniques, there is often only vague familiarity with the details of how they are implemented. This paper is intended to help alleviate this situation. Its purpose is to present an overview of GAs and NNs, which includes an explanation of what they are, how they work, and an example of how they are implemented.

Description: Recent articles addressed the subject of fuzzy regression [FR] and the term structure of interest rates [TSIR]. Their approach relied on possibilistic regression. Although possibilistic regression has been used in many applications, it has a number of limitations, not the least of which is its nebulous relation to the least-squares concept. As an alternative to possibilistic regression, this paper uses fuzzy least square regression to investigate the TSIR.

Description: Presented at October 1985 Annual Meeting. This session focused on three projects funded by the Actuarial Education and Research Fund: risk models, operations research applications for actuaries and probability modeling of losses.

Description: This 1978 session on classification models included 1. models demonstrating the effect of a classification system superimposed on a merit rating system and 2. models demonstrating the subsidization of poorer risks.

Description: This paper presents a history of traditional topics like actuarial cost methods and assumptions, the history of past inquiries into the dynamic and stochastic nature of pension costs, and the perceptions and concerns of pension actuaries of the past. From The Future of Pension Plan Funding and Disclosure Monograph, M-RS05-1.

Description: Presented at October 1986 Annual Meeting. This session discussion is about the use of simulation, or Monte Carlo, techniques to analyze complex insurance models such as estimating reserves, solvency considerations, and estimation of mortality and distribution of aggregate losses with the aid of a computer program

Description: A mathematical framework for benefits and choices must be created to model benefit selection. This paper creates such a framework by defining benefit plans as reimbursement functions. These are then used with a defined choice function to calculate the cost deviation due to selection. Finally, utility functions can be applied to this framework to predict choice.