Instruction

Name Cr Method of study Time Location Organiser
Bayesian inference in biosciences 5 Cr Lecture Course 29.10.2018 - 13.12.2018
Name Cr Method of study Time Location Organiser
Bayesian inference in biosciences 5 Cr Lecture Course 30.10.2017 - 14.12.2017

Target group

Master's Programme in Life Science Informatics is responsible for the course.

Module where the course belongs to:

  • Eco-evolutionary informatics

The course is available to students from other degree programmes. The

course is recommended especially for students in

* Doc­toral Pro­gramme in In­ter­dis­cip­lin­ary En­vir­on­mental Sci­ences (Denvi)

* Doc­toral Pro­gramme in Wild­life Bio­logy (Luova)

* Datascience Master's programme

Prerequisites

Basics in statistics and probability calculus.

Learning outcomes

The course covers the basic theory behind probabilistic and Bayesian modelling and their applications to common problems in environmental and biological sciences. After the The course students understand the Bayes rule and the related concepts, including prior, posterior and predictive distribution and likelihood function. Students will also be familiar with graphical model representation and basics in model assessment, criticism and comparison. Students are also able to apply Bayes rule to construct simple hierarchical Bayesian models. Students are familiar with the basic concept of Markov chain Monte Carlo (MCMC) and are able to apply MCMC methods to solve hierarchical Bayesian models using the JAGS/STAN software.

Timing

Recommended time for completion is during the first year of masters studies or the first year of PhD studies.

The course is offered in 2nd period.

Contents

  • Part 1: Introduction to Bayesian inference: Bayes rule, prior and posterior distribution, likelihood function, Binomial model, mark-recapture analysis
  • Part 2: Technical necessities and few practical models: Monte Carlo methods, Markov chain Monte Carlo (MCMC), marginalization, prediction, JAGS software, linear and generalized linear models
  • Part 3: Hierarchical models and Probability fundamentals: exchangeability, conditional independence, graphical models, hierarchical Binomial and Gaussian models
  • Part 4: Model assessment, criticism and comparison: posterior predictive check, sensitivity analysis, posterior predictive comparison, cross-validation, Bayes Factor

Activities and teaching methods in support of learning

Students are required to read course material, complete at least 50% of the exercises and pass the exam. Lectures are interactive and teacher will assist with exercises during the lectures and via Moodle between the lectures. Additionally, there is exercise group once a week where students can get help for exercises.

Study materials

  • Gelman, A., Carlin, J. B., Stern, H. S., Dunson D. B., Vehtari A. and Rubin, D. B. (2013). Bayesian Data Analysis. Chapman & Hall/CRC. Second or third edition.
  • Number of selected articles
  • R programming environment and JAGS/STAN software

The book is read partially (chapters to be announced during the course).

The articles need to be read fully.

Assessment practices and criteria

The course grade will be 0.5*[grade from the exercises] + 0.5*[grade from the exam]

Recommended optional studies

After the course students are recommended to continue to:

MAST32004, Advanced course in Bayesian statistics, 5 cr LSI35003, Project work in eco-evolutionary informatics, 5 cr 57429, Spatial modelling and Bayesian inference, 5 cr

Completion methods

The course consists of lectures, exercises and an exam. Completion requires solving at least 50% of the exercises and passing the exam.