Module Aims

This module provides a comprehensive introduction to geostatistical modelling and its applications in epidemiology. It aims to equip students with a robust understanding of the statistical modelling framework—spanning formulation, estimation, diagnostic evaluation, prediction, and scientific interpretation—tailored to geostatistical methods. Additionally, the course enhances skills in statistical analysis using the R programming environment, with a focus on specialized packages such as PrevMap and lme4. 

Module Learning Outcomes

By the end of this module students should be able to: 

1. Identify key characteristics of geostatistical problems.
2. Analyse regression residuals to detect spatial correlation.
3. Develop and fit stochastic models to geostatistical data exhibiting spatial correlation.
4. Create predictive maps of spatially varying phenomena.
5. Articulate how spatial statistical methods contribute to understanding scientific processes with spatial variation.

  Pre-requisites

• Basic knowledge of R programming 

• Solid knowledge of generalised linear models 

• Foundational knowledge of probability concepts, including density functions, cumulative density functions, quantiles, marginal and conditional distributions, and moments of random variables. 

 Teaching Strategy

The module combines lectures with practical sessions to introduce geostatistical methods and their applications. Real-world datasets, particularly from tropical disease and environmental epidemiology, will be used for hands-on exercises. 

 Assessment

The assessment involves a guided analysis of geostatistical data, culminating in a 1,500-word report. 

Module Length

4 days.