Description: In the insurance industry, machine learning models are increasingly used to make decisions that affect people's lives. In order to ensure that models are fair, they should be audited for bias. There are no hard and fast rules for bias audit; metrics that can be used to evaluate fairness vary depending on the context in which the model is deployed. In this session, we present a flexible framework that highlights key considerations in bias evaluation that can be adjusted to suit your process. In this session, we will learn about different metrics to assess fairness, discuss open source tools that can be used for bias audit, and walk through a case study.

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