PhD course in Aalborg

On 21-22 April 2015 Adrian Baddeley will give a two day PhD course at Aalborg University titled Analysing spatial point patterns with R. The course has a strong focus on hands-on exercises using the comprehensive R package spatstat. No prior knowledge about spatial statistics is required, but basic familiarity with R and ordinary statistical concepts are needed.

Course material:

The course is based on short blackboard lectures and software demonstrations followed by “lab”" exercises on the computer.

All the lecturer’s scripts (cheatsheets) and the source code for the labs are on:

Description:

Spatial point pattern datasets are becoming more common across many fields of research. However, statistical methodology for analysing these data has not been widely disseminated. This course is a practical introduction to the analysis of spatial point patterns with a strong focus on hands-on exercises throughout the course. The course gives an in-depth introduction to spatstat, which is an R-package for analysing spatial point patterns. The package supports a complete statistical analysis of spatial point pattern data: data input and inspection, calculations, plotting, exploratory data analysis, hypothesis tests, model-fitting, simulation, Monte Carlo methods and model diagnostics. The course will treat all these subjects, and in particular the goals are for the participants to:

  • Understand basic statistical concepts used in spatial point pattern analysis.
  • Get an overview of the capabilities of spatstat and how to find your way around.
  • Learn how to conduct a basic analysis of a point pattern dataset:
    • Calculating and plotting exploratory summaries.
    • Fitting Poisson, Cox, and Gibbs point process models.
    • Validating and critiquing fitted models.

To obtain these goals we:

  • begin with real examples and establish basic principles.
  • introduce relevant graphical/exploratory methods including non-parametric intensity estimation, pair correlation, Ripley’s K-function.
  • demonstrate model building, model-fitting, formal inference and model validation.

Prerequisites

  • Basic familiarity with R.
  • Basic familiarity with statistical concepts.
  • Please bring your own laptop, with the latest version of R and spatstat installed. Please ensure that all the ‘Suggested’ packages for spatstat are also installed.

Organizer: Assistant Professor Ege Rubak, e-mail: [email protected]
Lecturer: Professor Adrian Baddeley, Curtin University
ECTS: 1
Dates: 21-22 April 2015
Hours: 9:00-12:00 and 13:00-16:00 both days
Place: Aalborg University, Fredrik Bajers Vej 7G, Room G5-112
Number of seats: 40
Deadline: 1 April 2015