Health care research is moving toward analytic systems that take large health databases and estimate varying quantities of interest both quickly and robustly, incorporating advances from statistics, econometrics, and computer science. This webinar will discuss specific challenges related to developing and deploying statistical machine learning algorithms in prediction and causal inference for health policy research questions, including examples from the areas of health plan payment, mental health care, and medical devices. Considerations go beyond typical measures of statistical assessment, and include concepts such as dataset shift and algorithmic fairness. This webinar is the second in a series. To review the first webinar, “Principles of Machine Learning for Prediction,” by Speaker David J. Vanness, PHD click here to login to the ISPOR portal for the complimentary recording.
- Understand the shortcomings of standard parametric regression techniques for the estimation of prediction and causal effect quantities;
- Be introduced to the ideas behind machine learning approaches as tools for confronting high-dimensional data;
- Become familiar with the properties and basic conceptual implementation of machine learning for prediction and causal effect estimation; and
- Recognize the centrality of fairness and generalizability considerations in quantitative health policy research.