UBDC Summer Training 2017: Introduction to Areal Data Modelling using R

When:
Wednesday 30 August 2017
Time:
10:00 – 13:00 BST
Where:
Jura Teaching Lab, Level 4 Annexe, University of Glasgow Library, Hillhead Street, Glasgow G12 8QE
GET DIRECTIONS

Spatial data are an increasingly common form of data in many different applications.

For modelling, there are three main forms of spatial data, point process, areal and geostatistical.

This course focuses on areal data, which are characterised by the region of interest being partitioned into a set of non-overlapping areal units and the feature of interest is recorded at each areal unit. Applications might include the number of hospital admissions recorded at each datazone, or the number of crimes recorded at postcode level.

There are a number of different statistical models that can be used in such a context, dependent on the questions of interest. They have in common one feature, that we cannot assume that the observations are independent, so we will need to introduce and model forms of spatial dependence

The objectives of this course are to introduce some common spatial regression models, to use R in visualizing and modelling such types of spatial data.

By the end of the course, you will be understand different measures of spatial dependence, be able to fit, test and check spatial models and undertake statistical inference on the model output.

Course instructors

Francesca Pannullo and Charis Chanialidis, School of Mathematics and Statistics, University of Glasgow

Course duration

Half day (Wednesday 30th August 2017, 10:00am – 1:00pm)

Course location

Jura teaching lab, Level 4 Annexe, Glasgow University Library

Audience

Social scientists, students and practitioners.

Fees

  • £25 - For UK registered students
  • £35 - For staff at UK academic institutions, Research Council UK funded researchers, UK public sector staff and staff at UK registered charity organisations
  • £50 - For all other participants

Pre-requisite knowledge

Introduction to R and some previous knowledge of spatial data is preferred.

Course content

  • Read spatial data from a variety of data formats.
  • Visualise areal data.
  • Assess the degree of spatial dependence, including Moran’s I.
  • Development of spatial regression models (SAR and CAR), fitting and checking.
  • Drawing inferences from the model results.

JOINTLY FUNDED BY