- Department: Mathematics
- Credit value: 20 credits
- Credit level: M
- Academic year of delivery: 2024-25
- See module specification for other years: 2023-24
This module introduces students to setting up models, their estimation and comparisons for diverse response and explanatory variables, departing from the traditional assumptions of normality and linearity.
Pre-requisite modules
Co-requisite modules
- None
Prohibited combinations
Occurrence | Teaching period |
---|---|
A | Semester 1 2024-25 |
This module introduces students to setting up models, their estimation and comparisons for diverse response and explanatory variables, departing from the traditional assumptions of normality and linearity.
By the end of the module, students will be able to:
1. Demonstrate the unifying role of exponential families when studying the association between response and explanatory variables measured in diverse scales.
2. Perform maximum likelihood based inference for GLMs, including in the context of logistic and Poisson regression.
3. Provide descriptive statistics and graphical summaries of information contained in data from survival experiments in different types of studies.
4. Use estimation and hypothesis testing for inference for survival data.
5. Use the statistical programme R to perform data analysis in the GLM and survival analysis contexts.
6. (M-level) Carry out self-directed learning to undertake more complex and involved analyses in the context of GLMs.
Data that motivates this module may arise from a variety of fields from medicine to insurance and engineering, each with their characteristic challenges departing from the standard modelling tools. This module will extend your modelling skills from the limited umbrella of linear models to now encompass response data that may be following distributions (such as Binomial, Poisson, Gamma) which are part of the exponential family of distributions and may be non-linearly connected to a set of covariates. This module will also introduce you to modelling survival data, where observations are measured as time to event and often subjected to censoring, e.g. event times are not known completely.
The M-level students will be provided with some self-directed learning material on an advanced topic pertaining to generalised linear models or survival analysis, e.g. proportional hazards, which will be assessed in the exam.
Task | % of module mark |
---|---|
Closed/in-person Exam (Centrally scheduled) | 100 |
None
Task | % of module mark |
---|---|
Closed/in-person Exam (Centrally scheduled) | 100 |
Current Department policy on feedback is available in the student handbook. Coursework and examinations will be marked and returned in accordance with this policy.
Annette J Dobson, Introduction to Generalized Linear Models, Second Edition, Chapman and Hall.
Klein, J.P. and Moeschberger, M.L. (2003). Survival Analysis: Techniques for Censored and Truncated Data, Second edition, Springer.