Description: This paper is about a concept known as the 'rate of transmission of information' or, more simply, as the 'information rate.' The authors establish the equivalence of arbitrage-free economic systems and a generalized betting model in such a way that the interpretation of the information rate, as the maximum rate of increase of an investor's capital, is preserved. As a corollary to this result, they show that the difference in the information rate between models represents the achievable incremental return associated with the superior predictor model. From Actuarial Research Clearing House 1993, Vol. 3.

Description: Interview with John Houston, Advanced Analytics and Predictive Modeling practice leader at Deloitte.

Description: Society of Actuaries research efforts currently in development

Description: Presented at June 1986 Spring Meeting. This session discussion is about the National Association Of Insurance Commissioners' Early Warning System, A.M. Best Company ratios, Conning and Company solvency ratios, and California Department of Insurance Empirical ratios.

Description: This is a discussion for the paper by Manuel Gelles titled, 'Actuarial Considerations in Cost Analysis'. Commentators add their own perspectives and critiques.

Description: Interview with Ian Duncan about his actuarial background and role in predictive analytics.

Description: Abstract: Project portfolio risk management and risk analysis form one of the critical components of enterprise risk management. Organizations measure and analyze risks associated with projects, project portfolios, and programs. The important step of the process is quantitative risk analysis of project schedules using event chain methodology (ECM). ECM is a stochastic modeling technique for schedule risk analysis. All risks, including schedule and nonschedule-related risks, are assigned to a particular project and within this project to the particular activity or resource. Further, ECM allows one to model the relationship between project risks by defining risks that cause or trigger other risks. All risks and relationships between them will be presented on the project or portfolio Gantt charts using event chain diagrams. After risks are assigned to project and portfolio schedules, Monte Carlo simulation of the project schedule is performed. ECM simplifies complex risk analysis process.

Description: Here we break down deep learning into its various conceptual sets so as to have a better intuitive understanding of deep learning.

Description: Bayesian networks are one of many models that are useful for predictive modeling. This article introduces the concept of Bayesian network and shows how time-series data can be modeled for a Bayesian network.

Description: A discussion of permissible use and collection of genetic information for Health Risk Assessments and wellness programs under the Genetic Information Nondiscrimination Act of 2008.