ECS530 Analysing spatial data

• Topics

  This intensive course is addressed to doctoral and other students who need to learn applied spatial data analysis, most likely for thesis use, and to applied researchers with similar requirements; relevant disciplines include among others economics and econometrics, political science, ecology, epidemiology and public health, environmental science, applied statistics, geography.

• Learning outcome

  Upon completion of the course, the candidate ...

  Knowledge

  • will have knowledge of chosen internal and external representations of spatial data in R (and Python, based on existing familiarity with R, and Python if using Python), import, export, and transform the coordinate reference systems of spatial data, and be able to choose suitable visualisation methods
  • will be able to decide whether specifically spatial techniques are required in analysing spatial data
  • will be able to define spatial weights for testing for and modelling spatial autocorrelation and be able to test for spatial autocorrelation, and understand the assumptions underlying tests
  • will have an overview of spatial regression techniques, including spatial econometrics, and be able to apply them.

  Depending on modules chosen, the candidate ...

  • will have knowledge of problems encountered when deploying machine learning when using spatial data
  • will have knowledge of problems arising in R packages (Python modules) for handling spatial data as external software libraries evolve, especially for specifying and transforming coordinate reference systems.

  Skills

  • can formulate problems, plan and carry out spatial analysis in R and/or Python
  • can avoid and/or resolve problems related to programming spatial analysis in R and/or Python.

  Competence

  • can manage complex spatial data analysis assignments and projects in R and/or Python.

• Teaching

  Combined lectures and computer practicals, concluding with draft project presentations on final day. Participants use their own laptop computers. Autumn semester 2022: the talks will be streamed and recorded. It will be possible to participate remotely, but for those able to travel to Bergen, interaction with other participants will be more fruitful. Campus-based participation is strongly encouraged.

  The course is moving towards a module-based course with modules: I spatial data representation and visualization; II spatial autocorrelation and regression; III advanced topics in spatial data analysis. The modules are "flipped" based on recorded talks, materials and exercises, so campus or online interaction prior to the intensive week is to answer immediate questions and discussion topics raised by participants. The intensive week is in principle campus-based.

• Restricted access

  • PhD candidates from NHH
  • PhD candidates from University of Bergen
  • PhD candidates from other higher educational institutions
  • Promising master students if approved by course responsible
  • Researchers, post-docs and others for whom the course contents are relevant

• Recommended prerequisites

  Working knowledge of R and/or Python, and applied statistics/empirical analyses of quantitative data relevant for participants' disciplines, is expected; participants should prepare by refreshing these abilities if necessary before the course begins.

• Compulsory Activity

  Participation in preparation sessions (in-person/online), attendance (in-person/online during intensive week) and draft project presentation.

• Assessment

  Individual written project due 6 weeks in wiseflow after course ends. The project must be written in English.

  Compulsory activities (work requirements) is valid for one semester after the semester it was obtained. Re-take is offered the semester after the course was offered for students with valid compulsory activities (work requirements).

• Grading Scale

  Pass / fail.

• Computer tools

  R, additional contributed packages, RStudio, possibly a GIS (GRASS or other).

• Literature

  Bivand, Pebesma & Gómez-Rubio: Applied Spatial Data Analysis with R, Springer, 2013 (2nd edition) http://www.asdar-book.org/

  Lovelace, Nowosad & Muenchow (2019) Geocomputation with R https://geocompr.robinlovelace.net/

  Bivand, Pebesma (forthcoming) Spatial data science with applications in R - bookdown site, https://www.r-spatial.org/book.

  https://www.rspatial.org/

  Recommended reading: Information will be given later