The aim of the course is to provide the student with the fundamentals of model formulation, bayesian data analysis, and causal inference.

The course will consist of 10 x 4-hour sessions that will combine lectures to explain the theory and computer practicals to demonstrate analytical techniques. There will be 10 lectures and 7 practicals.

The students will be provided with the option to submit only 5 out of 7 practical reports (10% each) for this course. If more are submitted, they will be marked and can be used as formative assessment by the students. The best 5 marks will count towards their final mark. This flexibility is offered so that deadlines can be adhered to, but it means that no re-assessment opportunities will be offered for continuous assessment. They will then have to finish a final project (50%). The final project will be reassessed if necessary, as per university regulations.

The students will be provided with the option to submit only 5 out of 7 practical reports for this course. If more are submitted, they will be marked and can be used as formative assessment by the students. The best 5 marks will count towards their final mark. This flexibility is offered so that deadlines can be adhered to, but it means that no re-assessment opportunities will be offered for continuous assessment. The final project will be reassessed if necessary as per university regulations.

This course will provide the conceptual foundations needed to develop models tailored to specific questions and data in order to update existing knowledge. Emphasis will be on Bayesian approaches, using existing knowledge to build models of the data generation process and to estimate plausible parameter distributions. Students will be exposed to different computational methods to obtain parameter estimates and to assess model fit. Throughout the class, students will be made aware of to the importance of considering causal structures in model formulation and interpretation.

By the end of this course students will be able to:

1. Critically discuss key differences between a Bayesian and frequentist approach

2. Understand critically of how prior information is used in a Bayesian approach and their appropriate use

3. Understand and apply the concept of Markov Chain Monte Carlo techniques 

4. Critically discuss the importance of causal structures in model formulation and analysis

5. Write simple programs in JAGS and Stan

6. Critically evaluate and apply appropriate solutions to issues with model fit.

7. Build and interpret simple causal models using directed acyclic graphs

Students must submit at least 75% by weight of the components (including examinations) of the course's summative assessment.