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

Description: This article provides a brief overview of the current state of machine learning (ML) and artificial intelligence (AI) as they apply to actuarial science, including discussion around the integration of actuarial judgment with ML and AI. It concludes with an example of explainable ML through a python tool known as SHAP (SHapley Additive exPlanations), which is able to explain the key drivers behind individual ML or AI predictions.

Description: This paper outlines a statistical modeling framework for fitting mortality laws to mortality experience data. The process can be used as an alternative to traditional graduation techniques used in mortality table development.

Description: In this article we examine the benefits and drawbacks of using a global surrogate of a black-box model. Are surrogates a comfortable middle ground that provide insurers high predictive power as well as model explainability?

Description: In this paper, we illustrate how to value American-style options using the Least-Squares Monte Carlo LSM approach proposed by Longstaff and Schwartz 2001 and investigate whether there exists an optimal regression complexity in the LSM framework for options pricing. In particular, we use the smoothing spline in the regression step, which allows us to control the regression complexity on a continuous scale with just one tuning parameter. Numerical results on American put options indicate that we need to use more than a linear regression, but as the regression becomes more complex, the accuracy of the LSM method quickly deteriorates. From 39th Actuarial Research Conference.

Description: This paper outlines a method of parametric modeling that may be used to produce forecasts and forecast intervals that will convey a measure of the reliance that may be placed on the forecasts. Followed by 2 discussion papers submitted

Description: 1994 SOA Spring Meeting, Orlando. In this session the panel consider the question of uncertainty in loss reserves including discussion of sources and measurement. The panelists analyzed the same data set from different perspectives and discussed the strengths and weaknesses of the approaches.

Description: This article compares three potential approaches to determining new coefficients for health-based risk scoring models: full recalibration, p-value credibility weighting, and penalized ridge regression.

Description: An approach to quantify medical intervention savings when medical claims data is not available by leveraging overlapping measurement techniques.

Description: This article describes how different types of regression models used for fund modeling can affect the variable annuity's risk exposure report and attribution report.