Nov 17, 2019   1:52 p.m. Klaudia, štátny sviatok - Deň boja za slobodu a demokraciu
Academic information system

Course syllabus B1-ACR - Time series analysis (FCE - SS 2018/2019)


     Information sheet          ECTS          Syllabus          


     Slovak          English          


University: Slovak University of Technology in Bratislava
Faculty: Faculty of Civil Engineering
Course unit code: B1-ACR
Course unit title: Time series analysis
Mode of delivery, planned learning activities and teaching methods:
lecture2 hours weekly (on-site method)
seminar2 hours weekly (on-site method)

 
Credits allocated: 4
 
Recommended semester/trimester: Mathematical and Computational Modeling - bachelor (compulsory), 6. semester
Level of study: 1.
Prerequisites for registration: passed Statistical methods (B1-STM)
 
Assesment methods:
Accreditation:

After each lecture (with the exception of the last two) there will be a test from the theory covered by the lecture (for each of those tests a student can obtain 1 point at most, i.e. altogether no more than 10 points). These points together with the points of the programmes on exercises (maximum 25 points) shall be added to points from the exam. For course credit at least 60% (i.e., 21 points) are required.

Exam:

The test has two parts: the theoretical exam (a maximum of 30 points) and practical (project - a maximum of 35 points).
 
Learning outcomes of the course unit:
On completion of this course students will acquire knowledge of current methods of time series analysis, their decomposition by means of multidimensional regression, autocorrelation analysis, time series forecasting, model identification methodology, parameter estimates, model verification, and spectral analysis of one- and multi-dimensional time series. After the course completion a student should be able to analyze real time series, build their linear models and use them for description and prediction.
 
Course contents:
- Introduction to time series analysis.
- Decomposition of time series using multivariate regression.
- Autocorrelation analysis.
- Spectral anlysis. Application of spectral analysis in one- and multi-dimensional time series analysis.
- Box - Jenkins methodology: model identification (AR, MA, ARMA, ARIMA, SARIMA), parameters estimation, model checking.
- New models in Box - Jenkins methodology (ARCH, GARCH).
- Time series prediction.
 
Recommended or required reading:
Basic:
CIPRA, T. Analýza časových řad s aplikacemi v ekonomii. Praha : SNTL, 1986. 246 p.
ARLT, J. Moderní metody modelování ekonomických časových řad. Praha: GRADA Publ., 1999.
FRANSES, P. Time series models for business and economic forecasting. Cambridge: Cambridge University Press, 1998.

 
Language of instruction: slovak or english
 
Notes:
 
Courses evaluation:
Assessed students in total: 48

ABCDEFX
10,4 %6,3 %20,8 %22,9 %39,6 %0 %
Name of lecturer(s): doc. Ing. Tomáš Bacigál, PhD. (examiner, instructor, lecturer) - slovak, english
prof. RNDr. Magdaléna Komorníková, PhD. (lecturer, person responsible for course, tutor) - slovak, english
 
Last modification: 28. 2. 2019
Supervisor: prof. RNDr. Magdaléna Komorníková, PhD. and programme supervisor


Last modification made by Ing. Peter Korčák on 02/28/2019.

Type of output: