| Course title | Advanced regression methods |
|---|---|
| Course code | KBE/051 |
| Organizational form of instruction | Lecture + Lesson |
| Level of course | Doctoral |
| Year of study | not specified |
| Frequency of the course | In academic years starting with an odd year (e.g. 2017/2018), in the winter semester. |
| Semester | Winter |
| Number of ECTS credits | 5 |
| Language of instruction | Czech |
| Status of course | Compulsory |
| Form of instruction | Face-to-face |
| Work placements | This is not an internship |
| Recommended optional programme components | None |
| Lecturer(s) |
|---|
|
| Course content |
|
Content of lectures: LM, GLM and GAM, linear mixed-effect models with nested and crossed random effects, GLMM, GAMM, survival analysis with random effects, zero-inflated and zero-truncated models, regression trees and random forests, analyzing phylogenetical data, point pattern analysis, model selection and model averaging, bootstrap and jacknife methods. Content of practicals: Practicals complement the lecture, they are not temporally separated, but rather interleaved with it
|
| Learning activities and teaching methods |
Monologic (reading, lecture, briefing), E-learning
|
| Learning outcomes |
|
Students will learn to use the advanced methods of modelling the data from the field of natural sciences, with an accent on combining fixed and random effects in the models and properly choosing model complexity. An important component of this course are the individually solved homeworks, scored by the lecturer.
Students will be able to choose correctly - for datasets produced in the field of natural sciences - the type of regression models: coding of explanatory variables, choice of fixed and random effects, scale transformation for explanatory and response variables, testing significances of individual effects, methods of selection of explanatory variables. They will be also able to effectively apply advanced types of tree models (e.g. boosted regression trees, regression forests). They will manage the use of methods modelling phylogenetic correlations among taxa in the field of comparative ecology and to correctly use the non-parametric generalized additive models (GAM and GAMM). |
| Prerequisites |
|
The students should already know not only the basic types of linear models (linear regression and ANOVA], but also the more advanced generalized linear models (GLM) and the application of these models within the R software package. The ability of work with the R software is verified at the start of this course with a simple written test.
KBE/050 ----- or ----- KBE/785E |
| Assessment methods and criteria |
|
Written examination, Interim evaluation
Students should study the materials provided by lecturer before the corresponding topic is discussed in the lecture-practicals. Students are provided with homeworks (5-7 times) during the term, and their solutions are scored by the lecturer. |
| Recommended literature |
|
| Study plans that include the course |
| Faculty | Study plan (Version) | Category of Branch/Specialization | Recommended semester |
|---|