Course for international guest/part time students
- Faculty
- Faculty of Science
- Organization
- TTK Department of Probability Theory and Statistics
- Code
- infsta1u0um17em
- Title
- Information theoretic methods in statistics (l)
- Usual semester
- Spring
- Published semester
- 2025/26/2
- ECTS
- 3
- Language
- hu
- Learning outcomes
- Knowledge: getting familiar with the modern notions of information theory Ability: to understand and use information theory in statisitics Attitude: the need to extend the mathematical knowledge, to gain new analytic and applied programming skills Autonomy and Responsibility: based on the gained knowledge in information theory, the students are able to decide which tools are the most suitable to solve applied problems
- Course content
- Hypothesis testing: exponentially optimal trials for simple and complex null hypothesis testing, characterization of the optimal error exponent using I-divergence. The relationship between exponential distribution families, information projection and maximum likelihood estimation. Limit distribution of the maximum likelihood estimate. Analysis of contingency tables using the information theory method. The minimum description length method. Model selection based on information criteria
- Assessment method
- exam