Introduction to Bayesian Methods for Environmental Health and Beyond

This short course is a brief introduction to Bayesian inference and statistical modeling in R with applications for environment and health analyses. This one-day intensive workshop will cover the principles and practicalities of Bayesian inference, concepts like priors, posteriors, and non-normal probability distributions, and applications to regression modeling approaches and forecasting. Relevant real-world examples, such as modeling daily heat-related mortality to inform early warning thresholds will be used in the course. A blend of lectures and lab time will allow attendees to try out Bayesian methods and facilitate further exploration after the course.

105th AMS Annual Meeting
New Orleans Ernest N. Morial Convention Center
January 12, 2025 at 8:00 AM - 3:30 PM Central Time (In Person)

REGISTRATION RATES

REGISTER NOW

Course Description:

Bayesian inference and modeling are becoming more common in the health, climate, and meteorology worlds, but such methods are still not commonly taught as required courses in most academic or other training settings. As such, many researchers and practitioners will find themselves exposed to papers and analyses that relied on Bayesian methods without the skill to critically analyze and assess those approaches. Furthermore, as trans-disciplinary methods are also becoming quite commonplace, many experts in weather, water, and climate fields are collaborating with public health colleagues, and would benefit from understanding the language of biostatistics and becoming familiar with languages like R. This course addresses these shortcomings by providing a friendly and accessible starting point to empower such experts to engage more closely with both modern methods and colleagues in public health.

Participants will:

  • Understand the fundamental principles of Bayesian inference and contrast with other statistical methodologies such as frequentist approaches.
  • Be familiar with with the basics of setting up a Bayesian model for analysis in a modern programming environment.
  • Be able to run simple Bayesian models, explain the structure and approach to fitting them, and interpret the results of the models for applications like forecasting.

    VIEW AGENDA

    If you have questions regarding the course, please contact Hunter Jones.

    Instructors:

    Robbie Parks
    Robbie Parks

    Columbia University

    Akis Konstantinoudis
    Akis Konstantinoudis

    Imperial College London