Description: Futurism section members participated in this Delphi study, which had multiple choice questions on a range of life, health, economic and professional issues. The article provides background on the study and comments on the predictions of participants for 2010 (and 2050) in light of experience to date.

Description: Parables and Prophecies Prevent Predictive Prowess

Description: Alternative sources of learning available for free or at low cost if you wish to round out your F&F education about machine learning, artificial neural networks, genetic algorithms and the millennia of biological advances we draw from when we use these techniques.

Description: Discusses the factors one should take into account when considering the appropriate level of precision for a task or a subsequent communication.

Description: The purpose of this article is to provide a refresher on regression and times series models so that actuaries will feel more comfortable building forecasting models.

Description: This article is about research studies that use the Delphi technique. It describes what a Delphi study is and gives a thumbnail sketch of two Delphi studies the Society of Actuaries has sponsored in the past 10 years, as well as a recently initiated Delphi study.

Description: Read the April 2019 issue of Predictive Analytics & Futurism published by the Predictive Analytics & Futurism Section of the Society of Actuaries.

Description: Evaluates the ability of various model types to automatically approximate various mathematical formulas. This can be used as a guide for feature engineering.

Description: The well-known examples of predictive modeling “successes” are numerous. Perhaps less widely recognized is the myriad of ways in which a predictive model can fail to perform as expected, often due to misconceptions or misrepresentations on the part of the analyst. This article, which is intended for a general actuarial audience, focuses on three such pitfalls. By understanding these three common pitfalls, readers may guard against them within their own work, teams, and organizations.

Description: The newer predictive modeling tools are powerful but they work best when supported by a lot of computing power. Parallel processing is very powerful but often very difficult to implement. Dihui and Richard take us through the maze of options with H2O, SparkNet/MLib and Mahout. Which do you use for a generalized linear model (GLM), and which is better for high-volume data manipulation?