Bayesian Statistics I
This course will give you an introduction to Bayesian analysis, with emphasis on understanding the basic concepts and methods and simple problems that will be studied in detail. It will also give you an overview on how to analyse more complicated- and real-life statistical problems.
After taking this course, you will understand the difference between various interpretations of probability and be able to formulate a statistical problem on the basis of a Bayesian perspective. You will both learn to solve standard statistical problems using Bayesian methods and to solve statistical problems using simulation-based computational methods, such as the Markov Chain Monte Carlo (MCMC), which are often used in Bayesian inference.
In Bayesian inference, parameters are considered to be random variables and any previous knowledge about these parameters is expressed as a probability distribution, the so called a priori distribution. This prior distribution is then updated to a posterior distribution by using Bayes’ theorem to combine it with the observed data which is expressed through the likelihood function. The a posterior distribution, thus, expresses evidence about the parameters after data has been observed.
The course is given at day time, full time.
The course forms a part of the Master's Program in Statistics, but it can also be studied as a freestanding course.
The teaching forms consist of lectures and exercises.
More information for registered students will be found in Athena.
Examination will be in the form of written and oral examination.
Teachers autumn 2019