ST3201 Mastergradsseminar i statistikk – Spatial statistics

Notes from the organising meeting on September 1st

Course structure:

  • The seminar will be denoted by TMA4505 for Indmat students, and by ST3201 for others
  • The course is based on the spatial statistics course TMA4250, taught by Geir-Arne Fuglstad, and will reference video recorded lectures, exercises, textbooks, and lecture notes from the class on the wiki page (discussed in detail below) and Blackboard.
  • There are multiple textbooks for the class. Each lecture from the Spatial Statistics has associated reading from various textbooks.
  • The course will be divided into 3 parts:
  • Gaussian random fields (this will be the biggest part)
  • Point processes
  • Gaussian Markov random fields
  • There will be 7 meeting in total throughout the semester taking place every other week starting in week 37. Each meeting will be 1 hour long consisting of a 30 minute presentation and 30 minutes of discussion
  • Meeting times and places may vary throughout the semester
  • There will be a single required project based on the geostatistics portion of the class.
  • There will be an oral exam worth 100% of the grade. Details for the exam are not finalized, but it will probably be around 30 minutes, and will require knowledge of all topics throughout the semester.

Structure of individual meetings and presentations:

  • Each meeting will be 1 hour, with 30 minutes devoted to an individual student’s presentation, and 30 minutes devoted to discussion.
  • Presentations will summarize lectures from Geir-Arne’s Spatial Statistics class and associated reading from the textbooks, specifically the lectures and reading mentioned on the wiki page.
  • The presentations may also go through one of the exercises from the Spatial Statistics class associated with the lectures presented on.
  • The exact structure of the presentations is somewhat flexible, and may change throughout the semester.

Other • Johannes agreed to present for the first meeting, which will be on either Thursday or Friday in week 37 • Project groups will be 2-3 people

Meeting Date Responsible Lectures Lectures note Exercise Relevant Reading
1 16.09.23 Johannes Lecture 1.1 Slides, Lecture notes GG: Preface, 1.1
Lecture 1.2 Slides, Lecture Notes GG: 1.1, 1.2.1
2 29.09.23 Markus Lecture 2.1 Repetition, slides, Lecture notes Exercise, Solution GG: 1.3.1, 1.3.2, 1.3.3, 1.3.4, 1.4
Lecture 2.2 Repetition, slides, Lecture notes GG: 1.4, 1.8
3 13.10.23 Lecture 3.1
Lecture 3.2 part1,Lecture 3.2 part 2
4 26.10.23 PP1-Part1,PP1_Part2 Repetition, slides, Lecture notes Solution, Exercise [PP]: 1.1–1.9 (for motivation and overview), 2.1, and 2.3, [PP] 3.4.1, 6.2.1
PP2_part1, PP2_Part2 Repetition, slides, Lecture notes
5 03.11.23 Lecture Repetition, slides, Lecture notes [PP] 3.6.1–3.6.2, 6.3.2, 4.1–4.2 [You can read 3.4.2 and 6.4 if you are interested in more info on Cox processes]
Part 1, Part 2 Repetition, Lecture notes [PP] 4.2.2, 4.2.5–4.2.7, 4.3.1
6 10.11.23 Lecture Repetition, slides, Lecture notes Solution, Exercise [Tutorial]: 3.1–3.3
Part 1, Part 2 Repetition, Lecture notes [GMRF]: 1, 2.1.5, 2.2.1, 2.2.2.
7 24.11.23 Lecture Repetition, slides, Lecture notes [GMRF] 2.2–2.4
Part 1, Part 2 Repetition, slides, Lecture notes [GMRF] 3.3, 2.2.4


  • GG: Gaetan, C., & Guyon, X. (2010). Spatial statistics and modeling. New York: Springer.
  • PP: Illian, J., Penttinen, A., Stoyan, H., & Stoyan, D. (2008). Statistical analysis and modelling of spatial point patterns. John Wiley & Sons.
  • Tutorial: Hurn, M.A., Husby, O.K., & Rue, H. (2003). A Tutorial on Image Analysis. In: Møller J. (eds) Spatial Statistics and Computational Methods. Lecture Notes in Statistics, vol 173. Springer, New York, NY.
  • GMRF: Rue, H., & Held, L. (2005). Gaussian Markov random fields: theory and applications. CRC press.


Here is the project description and the data you need to use.

Some info about the project:

  • This project is compulsory. To be able to take the exam, a reasonable attempt must be made to solve all problems.
  • The project must be completed in groups of size 1–2.
  • The deadline is set to the end of November
  • The submission must be a nicely formatted and well-structured pdf-file. All figures must have captions, and be readable and easily-understandable. All figures must be described in the text. Use consistent notation for vectors, say \boldsymbol{…}, and matrices, say \mathbf{…}.
  • Submit code as separate files.
  • You can write names on the report
2023-11-13, Sara Martino