Applied Bayesian Statistics (April)

Organizing institution
Utrecht University - Faculty of Social and Behavioural Sciences
Course code
S64
Course fee (no housing)
€ 700.00
Level
Advanced Master
Apply now!

This 5-day course zooms in on the key concepts of Bayesian Statistics and advanced techniques for data-analysis. Topics that are covered include: density of the data, prior and posterior distribution, Gibbs sampling, the Metropolis-Hastings algorithm, the Bayes factor, the evaluation of informative hypotheses, use and elicitation of (informative) prior distributions, and data-analysis in R and JAGS. The course is aimed at researchers who not only work with statistical tools, but are also interested in the development and evaluation of statistical tools.

Bayesian statistics offers a flexible alternative to classical (frequentist) statistics. This course will provide a sound basis in Bayesian statistics for those who want to:

  • understand what Bayesian statistics is about;
  • use Bayesian statistics to build and evaluate statistical models;
  • get hands on experience with Bayesian statistics in R, JAGS, and Bain.

This 5 day course zooms in on the key concepts of Bayesian Statistics and advanced techniques for data-analysis. Topics that are covered include: density of the data, prior distribution, posterior distribution, the Bayesian p-value, Gibbs sampling, the Metropolis-Hastings algorithm, the Bayes factor, the evaluation of informative hypotheses, and using and eliciting (informative) prior distributions when estimating the parameters of, for example, correlation and growth curve models.

The course is aimed at researchers who not only work with statistical tools, but are also interested in the development and evaluation of statistical tools. Among these are psychometricians, sociometricians, epidemiologists, and statisticians. The only requirement is familiarity with the following concepts: the likelihood function, the p-value, analysis of variance, and multiple regression.

Note: Participants need to bring a laptop computer to the course, with RStudio (https://rstudio.com/), R ((https://www.r-project.org/), the R package bain, and JAGS installed (https://sourceforge.net/projects/mcmc-jags/files/JAGS/4.x/).

Among our Methodology and Statistics postgraduate courses, there is one other course that addresses Bayesian statistics. The distinction between these courses is as follows:

  1. Hypothesis Testing 3.0; an applied course on evaluating theory-based hypotheses (via Bayesian and information-theoretic model selection) and on addressing causes of the replication crisis. This course focusses on the application of model/hypothesis selection methods.
  2. (This course) Applied Bayesian Statistics, a course to learn more about Bayes’ theorem, Gibbs sampling and the Metropolis-Hastings algorithm, Bayes factors, the evaluation of informative hypotheses, and the use and elicitation of prior knowledge in Bayesian Estimation. This course pays attention to the statistical theory and formulas are being used.

For an overview of all our summer school courses offered by the Department of Methodology and Statistics please click here.

Course director

Prof. dr. Herbert Hoijtink

Lecturers

Prof. dr. Herbert Hoijtink, Dr. Ellen Hamaker, Dr. Caspar van Lissa and Dr. Mariëlle Zondervan-Zwijnenburg.

Target audience

The course is aimed at researchers who not only work with statistical tools, but are also interested in the development and evaluation of statistical tools. Among these are psychometricians, sociometricians, epidemiologists and statisticians. The only requirement is familiarity with the following concepts: the likelihood function, the p-value, analysis of variance, and multiple regression. The maximum number of participants for this course is 40. About half of these places are reserved specifically for Epidemiology students and IOPS students, the other half for other participants (first-come-first-served).

Course aim

The aim of the course is to introduce the main concepts of Bayesian statistics to researchers who want to apply, develop and evaluate Bayesian statistics. To provide a sound basis in Bayesian statistics for those who • Want to understand what Bayesian statistics is about • Want to use Bayesian statistics to build and evaluate statistical models • Want to get hands on experience with Bayesian statistics in R and JAGS.

Study load

Five days (09.00 – 17.00 hrs.) Students who want to obtain a grade and credit points (Epidemiology Master Students 1.5 EC, IOPS PhD students 2 EC) for this course need to attend the grading session on Friday. Certificates of attendance are available upon request.

Costs

Course fee
€ 700.00

Tuition fee for PhD candidates from the Faculty of Social and Behavioural Sciences from Utrecht University will be funded by the Graduate School of Social and Behavioural Sciences.

Scholarships

Utrecht Summer School does not offer scholarships for this course.

More information

Irma Reyersen | E: ms.summerschool@uu.nl

Recommended combinations
Hypothesis Testing 3.0

Registration

Application deadline: 27 March 2020