|
Lecturer(s)
|
-
Šmilauer Petr, doc. RNDr. Ph.D.
|
|
Course content
|
Content of lectures: Introduction to S language and R software (2 weeks), summarizing and extending linear models including ANOVA (3 weeks), generalized linear models (3 weeks), generalized additive models, classification and regression trees, survival analysis, linear mixed-effect models, the choice and combinations of regression models Content of practicals: Complements the lectures, not temporally separated, but rather inter-mixed with the lectures. Students also submit 5-6 homework tasks, scored by the lecturer.
|
|
Learning activities and teaching methods
|
Monologic (reading, lecture, briefing), Demonstration, Projection, E-learning
- Class attendance
- 56 hours per semester
- Preparation for classes
- 42 hours per semester
- Preparation for credit
- 16 hours per semester
- Preparation for exam
- 60 hours per semester
|
|
Learning outcomes
|
This course provides an alternative presentation of the basic types of statistical models and their extension by more advanced models. The focus is on the generalized linear models (GLM), but other model types are also introduced, both those related to LM/GLM (generalized additive models, mixed-effect models) and unrelated ones (survival analysis, classification and regressiong trees).
The students learn to: (1) work with R software on the Microsoft Windows platform, including data import from a table processor (e.g. MS Excel) (2) choose and use linear or generalized linear models for data analysis and perform regression diagnostic for fitted models (3) choose an appropriate method of the event analysis in time, typically as a survival analysis (4) properly apply the methods of regression or classification trees, depending on the type of response variable, and also to perform cross-validation to determine optimum tree size (5) work with the simpler types of linear and non-linear models with mixed effects
|
|
Prerequisites
|
Students already understand the principles of testing statistical hypotheses and know the basic types of linear models (including ANOVA models) in an extent matching the Biostatistika course (KBE/012)
KBE/012
|
|
Assessment methods and criteria
|
Written examination, Test, 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. Practicals are interleaved with lecturing, there is no separation in time. Participation in practicals is checked.
|
|
Recommended literature
|
-
J. Fox (2008): Applied regression analysis and generalized linear models. Sage.
-
J.M. Chambers & T.J. Hastie (1992, eds): Statistical models in S. Wadsworth.
|