Introduction to Bayesian Statistics

July 11-13, 2016 at NCAR

Course Description

This three-day course will provide an introduction to the theory and methods of Bayesian Statistics. Bayesian Statistics is an emergent area of Statistics applicable to many problems and especially relevant in the context of uncertainty quantification. The course will cover some simple one- and two-parameter problems to provide a flavor for the mechanics of the Bayesian approach. We will then discuss Markov Chain Monte Carlo (MCMC) methods for making inference using more complicated (higher dimensional) models. By the end of the course the partcipants will fit and evaluate models for a dataset of atmospheric CO2 concentrations taken from ice-core measurements. We'll use the open-source statistical software R [https://www.r-project.org/] and the open-source MCMC package that works seamlessly with R called Stan [mc-stan.org/]. The format of the course is hands-on and participants will use their own laptops.

Instructors

The lead instructor for the course is Alix Gitelman, Professor of Statistics at Oregon State University. She will be assisted by Nathan Lenssen (Columbia University), Pulong Ma (University of Cincinnati), Felipe Tagle (University of Newcastle), Doug Nychka (NCAR/IMAGe), Dorit Hammerling (NCAR/IMAGe) and others. Seats are limited to 12 participants and to apply, please visit the Bayesian Statistics Application link on the left hand-side of the page.

Monday, July 11, 2016, NCAR Mesa Lab - Damon Room

9:00 - 10:15

Session 1, Introductions and overview

10:15 - 10:45

Break

10:45 - 12:00

Session 2, CO2 example, priors (working in R)

12:00 - 1:00

Lunch, on your own, ML Cafeteria is cash only

1:00 - 2:15

Session 3, Picking priors (Brian Reich)

2:15 - 2:45

Break

2:45 - 4:00

Session 4, Introduction to MCMC

Tuesday, July 12, 2016, NCAR Mesa Lab - Damon Room

9:00 - 10:15

Session 5, Software setup and introduction to nimble

10:15 - 10:45

Break

10:45 - 12:00

Session 6, CO2 example (working in R, nimble)

12:00 - 1:00

Lunch, on your own, ML Cafeteria is cash only

1:00 - 2:15

Session 7, Bayesian Model Checking

2:15 - 2:45

Break

2:45 - 4:00

Session 8, Introduction to hierarchical models

Wednesday, July 13, 2016, NCAR Mesa Lab - Damon Room

9:00 - 10:15

Session 9, Timeseries basics and tools in R and nimble

10:15 - 10:45

Break

10:45 - 12:00

Session 10, Extending the COmodel in R and nimble

12:00 - 1:00

Lunch, on your own, ML Cafeteria is cash only

1:00 - 2:15

Session 11, Hierarchical Bayesian model inverse problem

2:15 - 2:45

Break

2:45 - 4:00

Session 12, Implementation of inverse CO2 model in R and nimble