Workshop: Using and developing algorithms for hierarchical models with NIMBLE, including MCMC and Laplace approximation-based methods
Date and time: Tuesday - Friday, June 16 - June 17, 2026. 9 am – 5 pm each day.
Location: University of Alaska Fairbanks Campus, Akasofu Building, Room 401.
2160 North Koyukuk Drive, Fairbanks, AK 99709
Instructors: Chris Paciorek (University of California, Berkeley)
Description:
The workshop will provide in-depth, hands-on training for using NIMBLE, including:
- Setting up models in NIMBLE.
- Overview of different MCMC sampling strategies and how to use them in NIMBLE, including Hamiltonian Monte Carlo (HMC).
- Writing new distributions and functions for more flexible modeling and more efficient computation.
- Tips and tricks for improving computational efficiency.
- Use of Laplace approximation-based methods including NIMBLE's new INLA-like nested posterior approximation.
- Using specialized model components, including Bayesian non-parametric distributions (based on Dirichlet process priors), conditional auto-regressive (CAR) models for spatially correlated random fields, and reversible jump samplers for variable selection (based on participant interests).
- Use of NIMBLE for specific ecological models (based on participant interests).
- An introduction to programming new algorithms in NIMBLE, including use of automatic differentiation (AD) in algorithms.
No prior experience with NIMBLE is required, but attendees should have a background in statistics and basic familiarity with R. More about NIMBLE: https://r-nimble.org
About the instructors
Chris Paciorek is an Adjunct Professor, as well as a research computing consultant, in the Department of Statistics at the University of California, Berkeley. His statistical expertise is in the areas of Bayesian statistics and spatial statistics, with primary application to environmental and public health research. He is co-PI and a core development team member of the NIMBLE software project, which provides a platform for implementing algorithms for hierarchical modeling. Website:
https://www.stat.berkeley.edu/~paciorek
Registration Rates
Meeting participants must register. The 2-day registration fee will cover the short course and also includes 2 catered lunches, coffee, and refreshments. In-person participants will be limited to 24 people. Online participants will not be capped.
General Registration: The general registration rate for the 2-day meeting is $400. The registration rate applies through June 8 (Monday).
Late Registration: After June 8, a late registration rate of $450 will apply for the meeting.
Online Only Option: An online Zoom conference option for those that don't want or can't attend in-person for $150. Lunch is not provided.
Students: contact the organizer below to request a student fee waiver and travel funds that are available on a case-by-case basis, subject to need and available funds.
Meeting Contact
Derek Arnold
darnold10@alaska.edu