[ 951STCOBAYK14 ] KV Bayes Statistics

Workload Education level Study areas Responsible person Hours per week Coordinating university
4 ECTS M1 - Master's programme 1. year Statistics Helga Wagner 2 hpw Johannes Kepler University Linz
Detailed information
Pre-requisites keine
Original study plan Master's programme Statistics 2015W
Objectives Students are familiar with the Bayesian approach to statistics and are able to perform a conjugate Bayesian analysis as well as Bayes inference using MCMC methods
Subject Basic concepts: Bayes' theorem, prior distribution, posterior distribution conjugate analysis

Bayesian inference: point and interval estimation, hypothesis testing, model choice (marginal likelihood, Bayes factor), Bayesian prediction, posterior predictive model checks, asymptotics

priors: natural conjugate priors in exponential families, improper priors, Jeffrey's prior

introduction to MCMC methods: Metropolis Hastings algorithm, Gibbs sampling, data augmentation

Bayes analysis of statistical models: linear regression models logit and ordinal logit model; finite mixture model, random effects models

Criteria for evaluation examples
written project report
Methods Lecture
Computer lab
Language English
Study material Hoff P.D. (2009). A first course in Bayesian statistical analysis.
Albert J. (2009). Bayesian computation with R.
Robert C. (2007). The Bayesian Choice.
Changing subject? No
Corresponding lecture 4MSBAKV: KV Einf├╝hrung in die Bayes-Statistik (4 ECTS)
On-site course
Maximum number of participants 40
Assignment procedure Assignment according to priority