Warning! The directory is not yet complete and will be amended until the beginning of the term.
300491 UE Time Series Analysis using modern statistical methods (2023W)
Continuous assessment of course work
Labels
ON-SITE
Registration/Deregistration
Note: The time of your registration within the registration period has no effect on the allocation of places (no first come, first served).
- Registration is open from Th 07.09.2023 14:00 to Th 21.09.2023 18:00
- Deregistration possible until Su 15.10.2023 18:00
Details
max. 10 participants
Language: German
Lecturers
Classes (iCal) - next class is marked with N
- Tuesday 10.10. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 17.10. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 24.10. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 31.10. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 07.11. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 14.11. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 21.11. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 28.11. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 05.12. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 12.12. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 09.01. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 16.01. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 23.01. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
- Tuesday 30.01. 16:45 - 20:00 Seminarraum 1.3, Biologie Djerassiplatz 1, 1.005, Ebene 1
Information
Aims, contents and method of the course
Assessment and permitted materials
1. Periodic questionnaires
2. Contributions during discussions (see above)
2. Contributions during discussions (see above)
Minimum requirements and assessment criteria
1. Quality of responses to the questionnaires
2. Quality of the discussion contributions.ad 1 (Questionnaires): 70%
ad 2 (discussion contributions) 30%passing grade: 60% or above
2. Quality of the discussion contributions.ad 1 (Questionnaires): 70%
ad 2 (discussion contributions) 30%passing grade: 60% or above
Examination topics
1. Defined by the material presented in the examples; more specific descriptions above.
Reading list
Key words in WIKIPEDIA
Lecture support material
Lecture support material
Association in the course directory
MAN 3
Last modified: Su 24.09.2023 20:28
The topics (see below) are introduced, based on statistical theory. Consequently, the practice interactions will be as follows: after the theory has been explained and perhaps elaborated, participants are challenged with questions how the theory is to be applied in specific cases. For example: given a set of described data, the students need to justify why or why not Bayesian analysis is to be applied.
Another example: given a time series of univariate measurements. Participants are asked which kind of smoothing algorithms can be applied; what the relative merits of the different smoothing algorithms are for the given data set.
Fallacies: (a) significance vs. significance tests; (b) mapping of categorical variables into integers; (c) correlation vs. regression; (d) standard error of the mean.
Data set comparisons: (a) comparisons of means; (b) maximum likelihood methods (Wilks lambda); (c) confusion matrices.
Smoothing: (a) window smoothing (mean, median, ...); (b) Laplace smoothing; (c) SVD (singular value decomposition) for multivariate data; (d) cumulative smoothing.
Dimension reduction: (a) Curse of Dimensionality; (b) (artifical) neural networks; (c) clustering.
Regression models: (a) polynomial regression (oLSq and GLM); (b) AICc (Akaike's Information Criterion with Takeuchi's correction for small sample size); (c) sigmoid, errorfunction, and arctangent regression; (d) issues of orthogonal bases functions (Fourier series.
Bayesian vs. Frequentist statistics: (a) Bayes Theorem/Rule; (b) conditional likelihoods; (c) Likelihood versus probability.