Výuka v letním semestru 2018/2019

Cvičení z Pravděpodobnostních a statistických problémů (NMSA160) - Úterý 10:40, 12:20, M2

Podmínky získání zápočtu: přiřazení na rozvrhový lístek v SIS, aktivní účast na cvičení, nejvýše 3 absence.

Materiály ke cvičení:
První cvičení (19.2.2019)
Druhé cvičení (26.2.2019)
Třetí cvičení (5.3.2019)
Čtvrté a páté cvičení (12.3. a 19.3.2019)
Šesté cvičení (26.3.2019)
Sedmé cvičení (9.4.2019)
Osmé cvičení (16.4.2019)
Deváté cvičení (30.4.2019)
Desáté cvičení (7.5.2019)

Cvičení z Prostorového modelování (NMTP438) - Pátek 10:40, K9

Podmínky získání zápočtu: aktivní účast na cvičení, přednesení alespoň tří referátů.

Materiály ke cvičení:
Zadání prvního bloku cvičení (náhodná pole na mřížích), ukázkový skript, ilustrace Isingova modelu.
Zadání druhého bloku cvičení (náhodná pole na souvislé množině).
Zadání třetího bloku cvičení.



Teaching in winter semester 2018/2019

Exercise class on Spatial Statistics (NMST543) - Monday 14:00, K10A

Requirements for obtaining the course credit: regular attendance, a short individual project.

Materials for the exercise classes:
Part I. (point processes, intensity)
Part II. (point processes, K-function)
Part III. (point processes, assessing interactions)
Part IV. (point processes, Monte Carlo and envelope tests)
Part V. (point processes, model fitting)
Part VI. (point processes on linear networks)
Part VII. (marked point processes)
Part VIII. (geostatistics, variogram)
Part IX. (geostatistics, kriging)
Part X. (random fields on a lattice)

Exercise class on Probability Theory 2 (NMSA405) - Monday 10:40, K8; Wednesday 14:00, K3

Requirements for obtaining the course credit: active participation (at most 3 absences), one homework presented at the blackboard.

Materials for the exercise classes:
Exercise sheets no. 1 and 2.
Exercise sheets no. 3 and 4.
Exercise sheets no. 5 to 7.
Exercise sheets no. 8 to 11.

Exercise class on Stochastic Processes 2 (NMSA409) - Tuesday 9:00, K5; Thursday 9:00, K5

Requirements for obtaining the course credit are given here. Note that the classes employ a variety of active learning strategies - some background material and reading is available here. Comments on the structure of the course are given here (in Czech).

Moodle for our course can be found here.

A collection of exercises with solution is available here (version 8.2.2018). English lecture notes are available here.

An example of the test assignment is available here.

The second exam takes place on Tuesday 8th January and Thursday 10th January. The second-chance test takes place on Thursday 17th January at 8:00 in the room K4.
Results of the second exam of the Tuesday class are here. For successfully passing the test at least 10 points out of 14 need to be obtained.
Results of the second exam of the Thursday class are here. For successfully passing the test at least 10 points out of 14 need to be obtained.

The first exam takes place on Thursday 15th November 2018. The second-chance test takes place on Thursday 29th November 2018 at 14:00 in the room K135 (KPMS).
Results of the first exam are here. We will have a meeting for discussing the results of the test and showing correct solutions on Thursday 22th November approx. at 14:10 in the room K135 (KPMS).

Recommended literature:
Prášková, Z.: Základy náhodných procesů II, Karolinum, Praha, 2007.
Brockwell P.J., Davis R.A.: Time series: Theory and Methods, Springer-Verlag, New York, 1987.

Materials for the exercise classes:
Block A, autocovariance function and stationarity.
Block B, L2-properties of stochastic processes.
Block C, spectral decomposition of the autocovariance function.
Block D, linear models of time series.
Block E, ergodicity, prediction.

Extra materials for the exercise classes (volunteers only):
Time series 1, filtration of the signal from noise, aim: obtain the clean signal.
Time series 2, prediction, aim: obtain predictions 1 and 2 steps ahead (using finite history of a chosen length, infinite history, estimating the prediction error based on different parts of the sequence, compare with the theoretical prediction error, if necessary fit a parametric model, check its validity).