The aim of this module is to introduce the student to the multiple linear and generalised linear regression models – the most widely used models in data analysis. Model formulation and interpretation will be explored in detail, including inclusion of categorical predictors and interactions.
Model building/identification techniques are critically examined. Methods for residual and influence diagnostics are covered. The R software system (or equivalent) will be utilised by the student as a tool for fitting these models
Multiple Regression Model
Motivation and formulation of the multiple regression model. Variance of parameter estimates and fitted values, confidence intervals and hypothesis testing. General linear hypotheses and ANOVA. Including categorical predictors in regression.
Model Building/Identification & model diagnostics.
Model building techniques, residuals and model diagnostics and their role in model appraisal.
Generalised Linear models: logistic & Poisson Regression
Logistic and Poisson regression models. Fitting GLMs with software. Interpretation of model parameters and other model output. General linear hypotheses of parameters. Wald's and likelihood ratio tests. Model building techniques.
A mix of live online classes, pre-recorded video lectures and live online software sessions.
| Module Content & Assessment | |
|---|---|
| Assessment Breakdown | % |
| Other Assessment(s) | 100 |