Computer Intensive Statistical Methods

Classical and Markov chain methods for stochastic simulation. Hierarchical Bayesian models and inference in these. The expectation maximisation (EM) algorithm. Bootstrapping, cross-validation and non-parametric methods. Classification.

1. Knowledge: The student knows computational intensive methods for doing statistical inference. This includes direct and iterative Monte Carlo simulations, as well as the expectation-maximisation algorithm and the bootstrap. The student has basic knowledge in how hierarchical Bayesian models can be used to formulate and solve complex statistical problems.

2. Skills: The student can apply computational methods, such as Monte Carlo simulations, the expectation-maximisation algorithm and the bootstrap, on simple applied problems.

3. General competence: The student is able to give an oral presentation within the subject area.

Lectures, works (projects) and student presentations of selected topics. Mandatory activities and the content of it will be communicated at the beginning of the semester. There is a final written exam.