- Student Records
Programme & Unit Catalogues

MA50247: Bayesian and large scale methods

Follow this link for further information on academic years Academic Year: 2019/0
Further information on owning departmentsOwning Department/School: Department of Mathematical Sciences
Further information on credits Credits: 6      [equivalent to 12 CATS credits]
Further information on notional study hours Notional Study Hours: 120
Further information on unit levels Level: Masters UG & PG (FHEQ level 7)
Further information on teaching periods Period:
Semester 2
Further information on unit assessment Assessment Summary: CW 40%, EX-TH 50%, OR 10%*
Further information on unit assessment Assessment Detail:
  • Coursework* (CW 40%)
  • Open Book Examination with a Duration of 24 hours* (EX-TH 50%)
  • Oral coursework presentation* (OR 10%)

*Assessment updated due to Covid-19 disruptions
Further information on supplementary assessment Supplementary Assessment:
MA50247 Coursework / Exam (dependent on failed component) (where allowed by programme regulations)
Further information on requisites Requisites: Before taking this module you must take MA40198
Further information on descriptions Description: Aims:
To introduce methods for large scale Bayesian stochastic modelling and statistical inference, including theoretical concepts as well as computational techniques.
To develop independent problem solving skills.

Learning Outcomes:
Students should be able to: formulate structured large scale Bayesian models; generate random samples efficiently from such models; analyse the structure of a Bayesian model in order to design a computational method for numerical inference; interpret and analyse the output of a Bayesian simulation or direct calculation algorithm.
To communicate: problem descriptions; model formulation; and inferences.

Problem solving (T,F&A), computing (T,F&A), written and oral presentation (F&A).

Bayesian modelling, inference and prediction.
Bayesian model assessment.
Large models and computational methods utilising sparsity and Markov properties.
Directed (hierarchical) and undirected graph models.
Designing efficient Markov chain Monte Carlo (MCMC) samplers.
MCMC output diagnostics.
Numerical techniques for direct, non-sampling Bayesian methods.
Further information on programme availabilityProgramme availability:

MA50247 is Optional on the following programmes:

Department of Mathematical Sciences