Úvod Výuka Studium Odkazy

## NMST434 Modern statistical methods (official information)

Requirements:

The knowledge of the statistics and probability theory at the level of courses Mathematical Statistics 1 and 2 (NMSA331 and NMSA332), Probability Theory 1 (NMSA333) and Linear regression (NMSA407).

Among others we will use the following concepts: almost sure convergence, convergence in probability, convergence in distribution, law of large numbers and central limit theorem for independent and identically distributed random vectors. A nice overview (in Czech language) of most of the results that are used in the course is available here.

(Assumed) content of the course:
• Asymptotic methods - Delta Theorem
• Theory of maximum likelihood
• Profile, conditional and marginal likelihood
• M-estimators and Z-estimators, Quasi-likelihood, Robust estimation
• Bootstrap
• Quantile regression
• EM-algorithm
• Methods for missing data
• Kernel density estimation
• Kernel nonparametric regression

Exercise classes
Exercise classes accompany the lecture with both theoretical as well as practical examples and illustrations.

Lecture material and actual information
Exercise class 3 (7. 3. 2017)
Exercise class 6 (28. 3. 2017)
Exercise class 7 (4. 4. 2017)
Exercise class 8 (11. 4. 2017)
• Working script - nmst434-E08.R
• Datasets - IQ-en.RData, (the code supposes that datasets are placed in the subdirectory `data' of the working directory)
Exercise class 9 (18. 4. 2017)
Exercise class 10 (25. 4. 2017)
• Working script - nmst434-E10.R
• AAUP.RData (the working script assumes that datasets are stored in the directory 'data' that is placed in the working directory)
Exercise class 11 (2. 5. 2017)
Exercise class 12 (9. 5. 2017)
Exercise class 13 (16. 5. 2017)
Exercise class 14 (23. 5. 2017)