|
Lecturer(s)
|
-
Voženílek Vít, prof. RNDr. CSc.
-
Hron Karel, prof. RNDr. Ph.D.
|
|
Course content
|
The content of the subject is adapted to the focus of the student's dissertation, it is assumed that the following topics will be selected in particular within object-oriented data analysis (multivariate data, compositional data, functional data, more complex data structures) according to the relevant geoinformatic motivation: - Linear and nonlinear regression and classification methods (including robust ones); - Model selection and evaluation (including cross validation, bootstrap); - Statistical inference (model parameter estimates, hypothesis testing, prediction; Bayesian approach); - Machine learning methods (regression / classification trees, support vector machines, neural networks); - Unsupervised methods (dimension reduction, cluster analysis, detection of outliers); - Methods for high-dimensional data (regression, classification, unreserved methods); Geostatistics, spatial and spatio-temporal statistics.
|
|
Learning activities and teaching methods
|
- Preparation for the Exam
- 40 hours per semester
|
|
Learning outcomes
|
To get an overview of modern statistical methods in geoinformatics.
Application Application of modern statistical methods in geoinformatics.
|
|
Prerequisites
|
Basic knowledge of applied statistics.
|
|
Assessment methods and criteria
|
unspecified
Specialized text and oral expert debate
|
|
Recommended literature
|
-
Bühlmann, P. van der Geer, S. (2011). Statistics for high-dimensional data. Springer, Heidelberg.
-
Filzmoser, P., Hron, K., Templ, M. (2018). Applied compositional data analysis. Springer, Heidelberg.
-
Chun, Y., Griffith, D.A. (2013). Spatial statistics and geostatistics. SAGE Publishing, Thousand Oaks.
-
James, G., Witten, D., Hastie, T., Tibshirani, R. (2013). An introduction to statistical learning. Springer, New York.
-
Ramsay, J.O., Silverman, B.W. (2005). Functional data analysis. Springer, New York.
|