BAYESIAN STATISTICS - 2017/8
Module code: MAT3003
SANTITISSADEEKORN N Dr (Maths)
Number of Credits
FHEQ Level 6
Module cap (Maximum number of students)
Overall student workload
Independent Study Hours: 120
Lecture Hours: 36
|Assessment type||Unit of assessment||Weighting|
|Examination||EXAMINATION - 2 HOURS||80%|
|School-timetabled exam/test||CLASS TEST||20%|
Prerequisites / Co-requisites
The module looks at the branch of statistics called Bayesian Statistics. It relies on subjective probability and looks at why this is extremely useful for modelling realistic problems. The module covers an introduction to Bayesian statistics, incorporating prior to posterior analysis for a wide range of statistical models. This shows the students an alternative approach to the Classical statistics that they have studied so far and looks at various statistical techniques that they have studied before and gives them a Bayesian approach.
introduce the rationale for, the main techniques of, and general issues in Bayesian statistics
apply techniques to standard statistical models, including exponential families and linear models
apply Bayesian approaches to estimation and testing
introduce Bayesian prediction
consider the role of decision theory
|Analyse the differences between the Bayesian paradigm and frequentist statistical methods||KC|
|Calculate the posterior and predictive distribution and related quantities||KC|
|Define hierarchical models and state and prove related theorems||KC|
|Demonstrate how models can be written in hierarchical form and calculate posterior quantities||KCPT|
|Explain the arguments for and against the Bayesian paradigm.||KC|
C - Cognitive/analytical
K - Subject knowledge
T - Transferable skills
P - Professional/Practical skills
Indicative content includes:
review of distribution theory
subjective probability and prior distributions – noninformative and conjugate
prior to posterior analysis
exponential families, sufficiency and conjugate priors
Bayesian estimation and hypothesis testing
application to linear models
approximate methods to estimation
elements of decision theory and comparative inference
Methods of Teaching / Learning
The learning and teaching strategy is designed to provide:
A detailed introduction to the theory behind, methodology and approaches used in Bayesian statistics
Experience (through demonstration) of the methods used to interpret, understand and solve problems in analysis
The learning and teaching methods include:
3 x 1 hour lectures per week x 11 weeks, with additional notes on white board to supplement the module handbook and Q + A opportunities for students.
(every second week) 1 x 1 hour tutorial replaces one of the lectures for guided discussion of solutions to problem sheets provided to and worked on by students during the tutorial.
The assessment strategy is designed to provide students with the opportunity to demonstrate:
· Understanding of and ability to interpret and manipulate mathematical statements.
· Subject knowledge through the recall of key definitions, theorems and their proofs.
· Analytical ability through the solution of unseen problems in the test and exam.
Thus, the summative assessment for this module consists of:
· One two hour examination at the end of Semester 2; worth 75% module mark.
· One 50 minute in-semester test; worth 25% module mark.
Formative assessment and feedback
Students receive written feedback via a number of marked coursework assignments over an 11 week period. In addition, verbal feedback is provided by lecturer at biweekly tutorial lectures.
Reading list for BAYESIAN STATISTICS : http://aspire.surrey.ac.uk/modules/mat3003
Please note that the information detailed within this record is accurate at the time of publishing and may be subject to change. This record contains information for the most up to date version of the programme / module for the 2017/8 academic year.