Description: The minimum distance estimator proposed allows for the selection of the best fitting family member without prior distributional assumptions. It is shown that this estimator is consistent and can be made either robust or asymptotically efficient. Its asymptotic distribution is derived and and illustrated with automobile insurance data. From the Actuarial Research Clearing House 1990 Vol. 1

Description: This article discusses the use of parametric distributions to model health care cost.

Description: The article provides a high-level, non-technical, introduction to artificial neural networks (ANNs) and deep learning with an application to risk management. The method relies on typical ANN components such as nodes, activation functions, and optimization through gradient descent.

Description: LTC insurers are beginning to implement population health programs to improve the lives of their policyholders before and during a claim. Measuring the financial impact of these programs can be difficult for a variety of reasons and multiple methodologies are used to evaluate treatments or programs. This article proposes two different methods of randomized control trials to evaluate the impact of these programs through a case study approach, along with additional potential considerations when evaluating effectiveness.

Description: This paper introduces a conjecture as to a method for estimating how much has to be added to the probability of ruin by n claims to obtain the probability of ultimate ruin. From the Actuarial Research Clearing House 1994 Vol. 2 .

Description: This paper develops the formula for estimation of multivariate copula. These results can be applied to the estimation of correlation coefficients. From the Actuarial Research Clearing House 1994 Vol. 2 .

Description: Summary of the SOA research project on PBR Simplified Methods. Describes the multi-risk scenario generator, central estimate assumptions, assumption objectivity measures, and regulatory demonstrations per VM-20.

Description: Feature article about the use of approximations.

Description: Traditional risk scoring models include only fixed (non-random) demographic and condition-based variables. By including the provider effect as part of the risk scoring model as a random effect, we can get better estimates of provider performance (incorporating practice patterns conditional on patient morbidity profile). This approach can be generalized to study other factors that are now masked in error variance (i.e. area, Social Determinants) but if modeled could give us better estimates of risk