ISPOR MEMBERS ONLY Educational Webinar: Principles of Machine Learning for Prediction

Tuesday, July 30, 2019
12:00PM EDT | 5:00PM BST |6:00PM CEST- 1 hour in duration
Click here for time zone conversion

Speakers: David J. Vanness, PhD 

This webinar will introduce learners to the basic principles of machine learning (ML) for prediction in health economics and outcomes research (HEOR). A second webinar by Professor Sherri Rose, this winter, will further explore principles of ML for causal inference in HEOR.

The webinar will begin by framing the prediction problem and drawing both connections and contrasts with the related problems of explanation and causal inference. It will then confront the “Iron Law of Prediction” – which characterizes the goals of prediction in terms of “minimizing loss” – or the consequences of making poor predictions. The Iron Law sets up an interesting tradeoff where we may tolerate some bias (systematic over- or under-prediction) in order to reduce the variance of predictions. The webinar will then explore the principles of “cross validation” and the “tuning” of algorithms to minimize loss. Using a simulated dataset, the presenter will apply several ML algorithms to illustrate the principles of “complexity penalization,” “bagging,” “random feature selection,” and “boosting.” The first webinar will conclude by considering an empirical application of boosting to construct propensity scores – a method of prediction that is used to improve causal inference for other parameters of interest.

This webinar series will use a combination of mild mathematical formalization (yes, there will be equations) and visualization. While advanced training in statistics, epidemiology, or econometrics is not required, a basic comfort with statistical concepts will help learners get the most out of the sessions. The R code used to generate and analyze the simulated dataset will be available for download.

What You Will Learn:

  • Describe the Iron Law of Prediction as a tradeoff between bias and variance of predictions to minimize the loss associated with making poor predictions
  • Describe how cross-validation allows analysts to minimize loss
  • Describe how the process of tuning ML algorithms is used to optimize their performance in minimizing loss
  • Describe how the bagging, random feature selection, and boosting can be used to minimize loss.

This is a Members ONLY webinar.

Members will receive an official member invitation via Email.

Not a member?

Join ISPOR Here


David J. Vanness, PhD
Professor of Health Policy and Administration
The Pennsylvania State University, USA


Chair, ISPOR Statistical Methods in Health Economics and Outcomes Research Special Interest Group

Please note: On the day of the scheduled webinar, the first 500 registered participants will be accepted into the webinar. For those who are unable to attend, or would like to review the webinar at a later date, the full-length webinar recording will be made available at the ISPOR Educational Webinar Series webpage approximately 2 weeks after the scheduled Webinar.

Reservations are on a first-come, first-served basis for all ISPOR members.

Your browser is out-of-date

ISPOR recommends that you update your browser for more security, speed and the best experience on Update my browser now