UNIVERSITY OF KENT
Confirmation that this version of the module specification has been approved by the School
Learning and Teaching Committee:
…15th December 2014…..…………………….(date)
SECTION 1: MODULE SPECIFICATION
1.
Title of the module
MA771 Applied Stochastic Modelling.
2.
School or partner institution which will be responsible for management of the module
School of Mathematics, Statistics and Actuarial Science.
3.
Start date of the module
revised version start date September 2015
4.
The number of students expected to take the module
30
5.
Modules to be withdrawn on the introduction of this proposed module and consultation with other
relevant Schools and Faculties regarding the withdrawal
None
6.
The level of the module (e.g. Certificate [C], Intermediate [I], Honours [H] or Postgraduate [M])
H
7.
The number of credits and the ECTS value which the module represents
15 credits (7.5 ECTS)
8.
Which term(s) the module is to be taught in (or other teaching pattern)
Spring term
9.
Prerequisite and co-requisite modules
Prerequisite modules: MA632: Regression, MA629: Probability and Inference.
There are no co-requisite modules.
10. The programmes of study to which the module contributes
BSc (Hons) Mathematics , BSc (Hons) Mathematics and Statistics, BSc (Hons) Financial
Mathematics, BA (Hons) Mathematics and Accounting & Finance (including programmes with a
Year in Industry), BSc (Hons) Mathematics with a Foundation Year
11. The intended subject specific learning outcomes
On successful completion of the module, students will:
a) Have revised and integrated the main probability and statistical material of a standard
undergraduate degree programme.
b) Have encountered a range of complex data.
c) Have an appreciation of probability models may be formulated for atypical data sets.
d) Have a good understanding of how likelihood-based classical procedures operate in practice.
e) Have experience of running a wide range of modern statistical procedures through running
computer programs in MATLAB.
12. The intended generic learning outcomes
On successful completion of the module, students will:
a) Appreciate the importance of computing for modern statistical analysis.
1
UNIVERSITY OF KENT
b)
c)
d)
e)
f)
Appreciate the breadth and importance of modern statistical methods.
Be able to describe a number of practical areas where statistical modelling is of importance.
Have enhanced their computer skills.
Have improved their ability to communicate effectively, and to work independently.
Have improved their skills in numeracy, problem solving, computing and written communication.
13. A synopsis of the curriculum
The lectures introduce students to a range of real-life data, and to ways of describing these data
sets by means of probability models. Statistical inference may then be classical or Bayesian. In
either case, it is necessary for the students to use computer programs to perform a statistical
analysis. The course is currently illustrated by about 50 MATLAB programs, which introduce the
students to many aspects of modern integrated computer languages. The course builds on the
statistics and probability results from earlier courses. The lectures illustrate how statistical
methods operated in practice, at the end of the 20th century.
Lecture syllabus:
Introduction and examples; introduction to MATLAB. Motivation through a range of real
examples, several taken from research in the school.
Model fitting by maximum-likelihood. Progression from explicit estimates to non-linear
problems. Importance of modelling through example results.
Function optimisation: deterministic and stochastic methods; simulated annealing.
Basic Likelihood Tools: profile likelihood; use of information criteria; score tests, Wald tests;
confidence interval construction.
Fundamental principles of modelling: Parameterisation: staying in range; delta method;
orthogonality; the EM algorithm; application to multinomials, censored data, mixtures, truncated
data.
Simulation techniques: simulating random variables; verification; kernel density estimation;
Monte Carlo inference; confidence interval construction; bootstrap; integral estimation;
importance sampling; Monte Carlo testing.
Generalised linear and generalised linear mixed models: exponential
family; fitting generalised linear models; analysis of deviance.
14. Indicative Reading List
Morgan, B. J. T. (2009) Applied stochastic modelling, Chapman and Hall.
McCullagh, P. and Nelder, J. A. (1989) Generalized linear models, Chapman and Hall.
15. Learning and Teaching Methods, including the nature and number of contact hours and the total
study hours which will be expected of students, and how these relate to achievement of the intended
module learning outcomes
Learning and Teaching Methods:
 32 hours of lectures, 8 hours of terminal classes.
 110 hours independent study
 Total number of study hours 150
Achievement of module learning outcomes:
 The lectures will address learning outcomes 11(a) -(d), 12(a)-(c),(e),(f).
 The terminal classes will address learning outcomes 11(b),(e), 12(d)-(f).
16. Assessment methods and how these relate to testing achievement of the intended module learning
outcomes
Assessment: The unit is assessed by examination (80%) and by continuous assessment (20%).
Continuous Assessment: This will consist of several open book written assessments started in
the terminal sessions and completed in independent study hours. These consist of questions on
numerical problems along with MATLAB computing problems and they test the learning
outcomes 11(b)-(e), 12(a),(b),(d)-(f).
Examination: A 2-hour written examination in the Summer term that will consist of questions and
numerical problems testing the learning outcomes 11(a)-(d), 12(a)-(c),(e),(f).
2
UNIVERSITY OF KENT
17. Implications for learning resources, including staff, library, IT and space
There should be no additional impact on the above resources.
18. The School recognises and has embedded the expectations of current disability equality legislation,
and supports students with a declared disability or special educational need in its teaching. Within
this module we will make reasonable adjustments wherever necessary, including additional or
substitute materials, teaching modes or assessment methods for students who have declared and
discussed their learning support needs. Arrangements for students with declared disabilities will be
made on an individual basis, in consultation with the University’s disability/dyslexia support service,
and specialist support will be provided where needed.
19. Campus(es) where module will be delivered:
Canterbury
3