Description: Often actuarial practitioners are faced with working with data that is less than ideal. The data may be observed with gaps in it, a model may suggest variables that are observed at different frequencies, and sometimes predictive analytic results are very fragile to the inclusion or omission of just a few observations in the sample. Data, particularly big data, are often "messy" and something must be done about it. What is the actuary to do about these very practical matters? These types of questions are what this presentation is intended to address. Concerns on missing data, outliers, nonlinearity, mixed-frequency, cross-correlations are illustrated using real datasets in the Asia-Pacific countries. Some possible solutions to the messy data problems will be discussed.

Description: Often actuarial practitioners are faced with working with data that is less than ideal. The data may be observed with gaps in it, a model may suggest variables that are observed at different frequencies, and sometimes predictive analytic results are very fragile to the inclusion or omission of just a few observations in the sample. Data, particularly big data, are often "messy" and something must be done about it. What is the actuary to do about these very practical matters? These types of questions are what this presentation is intended to address. Concerns on missing data, outliers, nonlinearity, mixed-frequency, cross-correlations are illustrated using real datasets in the Asia-Pacific countries. Some possible solutions to the messy data problems will be discussed.

Description: The Delphi method is a non-quantitative forecasting approach that is most useful when other forecasting techniques appear to have limited value. The Society of Actuaries (SOA) has had some impressive results from Delphi studies (even quoted in the Wall Street Journal for their uncanny accuracy on long-range predictions). Attend this session to learn about this effective, yet underutilized method of leveraging the wisdom of crowds.

Description: Uncertainty of a predictive model is a fact of life that many insurers could be overlooking at their peril without a framework for assessing it. The article provides a framework to understand the prediction uncertainty, which can be critical for the insurance operations using predictive models. Coin toss example explains model uncertainty and randomness inherent in making predictions to the number of heads expected. A mortality analysis case study illustrates how the Bayesian methodology can be used to develop the prediction uncertainty. In particular, the article shows that the Bayesian framework is a way to combine historical data and actuarial judgment, and directly address the model uncertainty through its posterior distribution.

Description: This practical session would provide a technical overview about GLM techniques and describe how GLM can be applied as a business tool in solving business problems in life insurance. The session would also focus on recent real-world case-studies across different functional areas like pricing, fraud detection, underwriting and product development.

Description: This practical session would provide a technical overview about Generalized Linear Model (GLM) and describe how GLM can be applied as a business tool in life insurance. The session would also focus on case-studies sharing across different functional areas such as pricing, fraud detection, underwriting, product development and other real-world applications.

Description: This session is a primer on how to apply Gestalt psychology to display quantitative information to improve neural receptiveness. While much of brain functioning is still a mystery to psychologists and neuroscientists, we do understand a great deal about which parts of the brain engage or disengage given visual cues. Actuaries are often tasked with presenting very technical information and are prone to focus more on the substance of numerical data rather than a pictorial presentation based on neural preferences. This presentation will discuss how to display quantitative information more effectively tailored to these preferences.

Description: Deep learning neural network applied to experience analysis for post-level term lapse and mortality.

Description: We compare two predictive analytics methods - GLM and GBM in terms of predictiveness vs. interpretability.

Description: The objective of this paper is to provide a practical guide with concrete case studies to help actuaries implement artificial intelligence and machine learning (AIML) to help accelerate the speed of analysis for life and annuity actuarial modeling. We do so by researching the existing literature regarding AIML and its applicability to stochastic modeling and exploring case studies.