Bayesian Modeling for Environmental Health Workshop

This training is essential for those who are interested or who have heard about Bayesian modeling and work in Environmental Health, but who have little theoretical or practical experience of it and would like some tools and know-how to get started in an approachable and friendly setting.

This three-day intensive workshop introduces the ideas of Bayesian inference and modeling in the context of Environmental Health, designed to be as approachable and friendly as possible while still providing technical and practical know-how. Led by a team of scientists with many years of diverse combined experience, the workshop will integrate seminar lectures with hands-on computer sessions to put concepts into practice. Several examples will be given using existing data, and conversations on starting new investigations with attendees' research questions will also be encouraged. The lectures and lab sessions will give an overview of the principles of Bayesian inference, as well as how to deal with different data structures, the various software options available, different types of analyses, and current and future research.

By the end of the workshop, participants will be familiar with the following topics:

•  Principles of Bayesian inference
• Practicalities of Bayesian inference
• Choosing priors
• Different data structures (spatial, point, continuous, categorical)
• Advantages and drawbacks of Bayesian approaches
• Temporal modeling
• Spatial modeling
• Spatio-temporal modeling
• Hierarchical modeling
• Exposure response functions
• Examples of use
• Software options

Summer 2026: The Bayesian Modeling for Environmental Health Workshop is a live, in-person training taking place August 5-7, 2026 from ~9:00am - ~5:00pm at the Columbia University Irving Medical Campus in NYC. All training start and end times are in EDT.

More information on travel, lodging, and getting around NYC.

Investigators at all career stages are welcome to attend, and we particularly encourage trainees and early-stage investigators to participate. There are three requirements to attend this training:

1. Basic familiarity with R and R studio (how to download R/R studio, and how to install a package) is recommended to get the most out of the workshop. 
2. Familiarity with spatial and temporal data structures, as well as exponential family distribution types (normal, Poisson etc.) would also be useful, though not essential.
3. Each participant is required to bring a personal laptop as all lab sessions will be done on your personal laptop. Each participant will be using RStudio Cloud to carry out tasks while attending the Workshop. Instructions for the basics of RStudio Cloud.

Knowing basic R platform and commands is required for the Workshop as noted in the requirements above. This training will use Posit Cloud (formerly RStudio Cloud). If you are new to R or need a refresher, you can review the below tutorials to be well prepared:

• R Programming Tutorial - Learn the Basics: A free datalab.cc class on R fundamentals
• Once you create your free, basic RStudio Cloud account for the training: Primers on Programming Basics and Visualization Basics.
• SHARP Program RStudio Cloud Tutorial: This self-paced tutorial from the Columbia SHARP Program walks through the Cloud Platform you will use at the training, as well as some basic exercises. We recommend this tutorial if you have not used the Cloud version of RStudio before, or if you are a beginner user of R.

If you have any specific questions about R and R studio in the context of the Bayesian Modeling for Environmental Health Workshop, please email us.

Other Software Introductions:

• Bugs, OpenBugs, and WINBugs
• RStan
• R-INLA
• R-NIMBLE
• NumPyro

Summer 2026 instructing team is being finalized, but will be comparable to the 2025 lineup below.

Training Co-Director: Robbie Parks, PhD, Mailman School of Public Health, Columbia University.

Training Co-Director: Garyfallos Konstantinoudis, PhD, Department of Epidemiology and Biostatistics, Imperial College London.

Training scholarships are available for the Bayesian Modeling for Environmental Health Workshop.

"The Bayesian Modeling Workshop is a very good introduction for spatio-temporal modeling in Public Health. It begins with the principal fundamentals in bayesian statistics and finishes with the most recent models. I recommend this course for any person interested in learning about statistical methods with spatial and temporal data." - Assistant Researcher at Pontificia Universidad Católica de Chile, 2025

"This was a great course with wonderful instructors and true professionals who are not only highly knowledgeable but also excellent teachers. The course was very well-structured, covering a wide range of material, starting from the basics and progressing to more complex cases. It was a valuable experience from which I learned a great deal." - Postdoc at University of Minnesota, 2025

"The workshop substantially improved my understanding of using bayesian models for environmental health in just two days given the real-world applicability of the workshop materials. The instructing team were very passionate and knowledgeable about the material." - PhD Candidate at University of Michigan School of Public Health, 2025

"A thorough and highly applicable introduction to Bayesian modeling." -  Postdoc at Icahn School of Medicine at Mount Sinai, 2024

"This course was an exercise in Bayesian praxis. It provided tools to implement theories I had only read about but couldn't figure out how to implement with code." - Faculty Member at Northern Arizona University, 2024

"This was a compelling workshop taught by people who know their material. Thorough without getting bogged down in the weeds, and practical examples are provided with pre-written scripts to follow." - Postdoc at Uniformed Services University of the Health Sciences, 2024

"The workshop was a meticulously prepared, and every detail was thoughtfully addressed. It was an inspiring experience to have the opportunity to learn from world-leading experts in the field." - Postdoc at Columbia University, 2023

"This is an efficient, in-depth course with lecture/lab pairs on topics including Bayesian workflows, temporal models, spatial models, and and non-parametric models. The course gives hands-on practice with the tools used for these analyses, with a focus on NIMBLE, but opportunity to work with INLA and ensemble models." - Student at University of North Carolina Chapel Hill School of Medicine, 2023

"This course was a great overview of how Bayesian models can be used within the environmental health/ public health arenas. User friendly with very knowledgeable instructors that made me excited to incorporate Bayesian models into my research!" - Postdoc at University of South Carolina, 2023

Registration Fee includes course material, breakfast, and lunch on training days. Course material will be available to all attendees during and after the workshop. Lodging and transportation are not included.

2026 Registration Category Rates: 

• ​​​Student/Postdoc/Trainee:
  • Early-bird rate: $1,395
  • Regular rate: $1,595
• Faculty/Academic Staff/Non-Profit Organizations/Government Agencies:
  • Early-bird rate: $1,595
  • Regular rate: $1,795
• Corporate/For-Profit Organizations:
  • Early-bird rate: $1,795
  • Regular rate: $1, 995

$200 early-bird discount is automatically applied if you register before the June 15 deadline.  

• $200 Early-bird Discount: This is automatically applied if you register before the June 15 early-bird deadline.  

• 10% Columbia Discount: This is valid for any active student, postdoc, staff, or faculty at Columbia University. If paying by credit card, use your Columbia email address during the registration process to automatically have the discount applied. If paying by internal transfer within Columbia, see below.  

• 10% Mailman Alumni Discount: This is valid for any individual who graduated from the Columbia University Mailman School of Public Health. To access the Mailman Alumni discount and receive a registration code, please email sharp_program@cumc.columbia.edu your graduation year and degree.  

• Group discounts are available for organizations sending 5+ participants. Please contact us directly at sharp_program@cumc.columbia.edu for more information.  

If paying by internal transfer within Columbia, submit this Columbia Internal Transfer Request form (link to form coming soon) to receive further instructions. Please note: filling out this form is not the same as registering for a training and does not guarantee a training seat.

If you would prefer to pay by invoice/check, please submit this Invoice Request form (link to form coming soon) to receive further instructions. Please note: filling out this form is not the same as registering for a training and does not guarantee a training seat.

Cancellation notices must be received via email at least 30 days prior to the training start date in order to receive a full refund, minus a $75 administrative fee. Cancellation notices received via email 14-29 days prior to the training will receive a 75% refund, minus a $75 administrative fee. Please email your cancellation notice to Columbia.Bayesian@gmail.com. Due to workshop capacity and preparation, we regret that we are unable to refund registration fees for cancellations less than 14 days prior to the training.  

If you are unable to attend the training, we encourage you to send a substitute within the same registration category. Please inform us of the substitute via email at least one week prior to the training so we can include them on attendee communications, gather registration details, and provide materials. Should the substitute fall within a different registration category (e.g., you are a faculty member and they are a postdoc), the credit card on file will be credited/charged respectively. Please email substitute inquiries to Columbia.Bayesian@gmail.com. In the event Columbia must cancel the event, your registration fee will be fully refunded.