The course consists of three modules with associated lecture notes, computer exercises and projects. Other useful links include:

An introductory lecture (L00-1x3.pdf) was given on the Image analysis (FMA170) course on 2019-09-26.

Examination

Examination consists of three home assignments/projects. The projects will be handed out during the 2^nd, 4^th and 6^th course week and reports are due 2 weeks later. The third project should also be presented at a short (10 minute) seminar either before or after Christmas. Presentation times will be arranged in consultation with the students and announced later.

For questions regarding the projects please use the discussion forums.

The computer exercises constitute good preparation for the projects; they are not mandatory.

Staff

Lecturer: Johan Lindström, MH:319.
- Office Hours: Fridays 15.00–16.30 (not first week).
Lab Assistant: Adrian Roth
Course Administrator: Susann Nordqvist, MH:221.

Literature

Most of the material is covered in two E-books available from the Lund University Libraries:

A. Gelfand, P. Diggle, P. Guttorp, M. Fuentes (Eds), Handbook of Spatial Statistics.
M. Blangiardo, M. Cameletti, Spatial and Spatio-temporal Bayesian Models with R-INLA.

Useful formula for matrix manipulations can be found in:

K. Petersen, M. Pedersen, The Matrix Cookbook.
The following chapters are of special relevance:
- 6.2 Expectation of Linear Combinations
- 8.1 Gaussians: Basics
- 9.1 Block matrices
- 9.6 Positive Definite and Semi-definite Matrices

Given previous CEQ comments regarding the literature the following books provide alternative reading. For books available as E-books from Lund University Libraries links have been provided:

P. Diggle, P. Ribeiro, Model-based Geostatistics. (covers the first module and parts of the second)
M. Stein, Interpolation of Spatial Data. (covers the first module)
N. Cressie, Statistics for Spatial Data. (a "classic" that covers the first module, good as a reference but slightly inaccessible)
S. Banerjee, B. Carlin, A. Gelfand, Hierarchical modeling and analysis for spatial data. (provides a slightly different approach to spatial data compared to the approach used in this course)
E. Krainski, V. Gómez-Rubio, H. Bakka, et al., Advanced spatial modeling with stochastic partial differential equations using R and INLA. (covers the second module, specifically chapter 2.2)
H. Rue, L. Held, Gaussian Markov random fields : theory and applications. (covers part of the second module, but does not contain the latest results)

Prerequisites

A basic course in mathematical statistics or one of:

Lectures will assume knowledge of Stochastic Processes and familiarity with Matlab is strongly recommended.

Recommended related courses (not prerequisites)

Monte Carlo methods for stochastic inference (FMSN50/MASM11)
Linear and Logistic Regression (FMSN30/MASM22)
Computer vision (FMA270)
Master thesis proposals

FMSN20/MASM25 - Spatial Statistics with Image Analysis