Description: Predictive modeling continues to gain traction in the life insurance industry with the rapid growth of available data and maturity of open source tools. New methods and technology in predictive analytics have enabled the evaluation of extremely large data more quickly and allow us to incorporate features previously not considered. The presenters will provide background on predictive modeling tools and algorithms utilized by actuaries and data scientists for life insurance, highlighting applications from marketing to underwriting.

Description: The presenters will discuss predictive analytics as applied to recent group life and disability experience studies and other applications.

Description: The management of the investments of an insurance enterprise has undergone a considerable evolution since the pioneering work of Redington in the mid 20th century. Best practices have evolved to incorporate ever more complex insurance liabilities, expanding investment asset classes and global market opportunities. Strategic asset allocation (SAA), liability driven investing, and other robust optimization approaches form the core of the professional management of insurance company investments.

Description: Institutional investors such as pension plans and insurance companies have enjoyed strong equity returns for a length of time with few historical precedents. Nearly a decade after the end of the financial crisis, strategies to provide downside protection remain expensive and the performance of traditional risk-managed funds has disappointed investors.

Description: In this session, we will discuss the advantages of using stochastic simulation of decrements in actuarial projections. We will also discuss practical approaches for parameterizing assumption probability distributions for simulation using predictive models as well as challenges the practitioner may face in implementation.

Description: Neural Networks are no longer limited to classification and regression solutions to numeric problems. CNNs add the ability to classify images, and GANs can create new images. Learn how these capabilities can provide competitive advantages to early adopters, how they work, and how to implement them.

Description: LON-CAPA is a full-featured course management, learning content management, and assessment system. It was initially developed by Michigan State University under several National Science Foundation grants. You can run LON-CAPA for free on your own dedicated server. Purdue University has used LON-CAPA to create a robust online homework system for the Financial Mathematics material of the FM Exam. This homework system is in the public domain and may be used by any university running LON-CAPA. This session will provide additional information and include a demonstration of the system.

Description: The optional phase-in period for life principle-based reserves (PBR) has come to a close, and companies have put in the work to bring their PBR implementations across the finish line. With the mandatory PBR implementation deadline behind them, companies can turn their attention from compliance to model and process improvements. The presenters will outline lessons learned from PBR implementations and their applications to broader use cases, as well as resources that are available for PBR modelers. Discussion topics will include items such as: Model simplifications and validation Model run-time techniques Model scalability (e.g. adding new products / future assumption updates) Sensitivity testing Model projections Available PBR & other modeling resources (e.g. PBR practice note, ASOP 56)

Description: The speech briefly analyzed the national reimbursement list from the perspective of health economics. The overall principle and logic behind it, and the relevant important considerations; In the construction and practical use of the national reimbursement list, it discussed the commercial insurance directory importance and enlightenment of national reimbursement list to establishment of commercial insurance directory.

Description: Many quantities arising in non-life insurance depend on claim severity distributions, which are usually modeled assuming a parametric form. Obtaining good estimates of the quantities, therefore, reduces to having good estimates of the model parameters. However, the notion of 'good estimate' depends on the problem at hand. For example, the maximum likelihood estimators [MLE] are efficient, but they generally lack robustness. Since outliers are common in insurance loss data, it is therefore important to have a method that allows one to balance between efficiency and robustness. This presentation suggests a general estimation method that we call the method of trimmed moments [MTM]. This method is appropriate for various model fitting situations including those for which a close fit in one or both tails of the distribution is not required. The MTM estimators can achieve various degrees of robustness, and they also allow the decision maker to easily see the actions of the estimators on the data, which makes them particularly appealing.