Further Statistics for Actuarial Science - MAN00065H
Module summary
The aim of the module is to expose students to a number of advanced statistical topics that are used in actuarial science and quantitative risk management, including Bayesian inferential procedures, credibility theory, extreme value theory, the modelling of dependent risks and machine learning
Module will run
Occurrence | Teaching period |
---|---|
A | Semester 1 2025-26 |
Module aims
The aim of the module is to expose students to a number of advanced statistical topics that are used in actuarial science and are part of the professional syllabus, including generalised linear models (GLMs), machine learning, extreme value theory, the modelling of dependent risks and mortality projection models.
Module learning outcomes
After successful completion the student is able to:
Subject content
-
explain the main theory and concepts of generalised linear models (GLMs);
-
apply GLMs to data and demonstrate knowledge of their importance in actuarial applications;
-
demonstrate knowledge of the main concepts of machine learning and explain their relevance to actuarial science;
-
demonstrate understanding of the main ideas of extreme value theory and apply these to actuarial loss modelling problems;
-
explain how dependence may be modelled at a deeper level than correlation and apply copulas to dependence modelling problems;
-
apply mortality projection models to data including the Lee-Carter model
Academic and graduate skills
-
present statistical analyses in a logical, rigorous, and concise way.
-
strict logical reasoning from assumptions to conclusion;
-
critically assess assumptions necessary to draw certain conclusions.
Module content
Syllabus:
-
Generalised linear models (GLMs)
-
Mortality projection
-
Introduction to machine learning
-
Extreme value theory (EVT)
-
Copulas and dependence
Indicative assessment
Task | % of module mark |
---|---|
Closed/in-person Exam (Centrally scheduled) | 70 |
Essay/coursework | 30 |
Special assessment rules
None
Indicative reassessment
Task | % of module mark |
---|---|
Closed/in-person Exam (Centrally scheduled) | 70 |
Essay/coursework | 30 |
Module feedback
Students will receive feedback within three weeks of the hand-in problem sets. The feedback will be handed to students personally and takes the form of comments and suggestions for improvement written on the handed in work.
Indicative reading
McCullagh and Nelder (1989), Generalized Linear Models (2nd ed), Chapman and Hall/CRC
Friedman, Tibshirani, Hastie (2009), The Elements of Statistical Learning: Data Mining, Inference and Prediction, Springer Verlag
McNeil, A, Frey, R and Embrechts, P (2016), “Quantitative Risk Management: Concepts, Techniques & Tools” (2nd ed), Princeton University Press.