Advanced time-series analysis
General data
Course ID: | 2400-M1IiEZASC |
Erasmus code / ISCED: |
14.3
|
Course title: | Advanced time-series analysis |
Name in Polish: | Zaawansowana analiza szeregów czasowych |
Organizational unit: | Faculty of Economic Sciences |
Course groups: |
(in Polish) Przedmioty kierunkowe (obowiązkowe) do wyboru - studia II stopnia IE - grupa 2 (3*30h) (in Polish) Przedmioty obowiązkowe dla I r. studiów magisterskich drugiego stopnia - Informatyka i Ekonometria |
ECTS credit allocation (and other scores): |
4.00
|
Language: | Polish |
Type of course: | obligatory courses |
Short description: |
The course offers practial workshops with R and RMarkdown languages. Practical examples of modern time series analysis will be presented: Box-Jenkins procedure, ARIMA/SARIMA models, ECM/VAR/VECM models, univariate and multivariate models from GARCH family |
Full description: |
1. introduction to R 2. stationarity, random walk, stochastic trends, stationarity testing, spurious regressions, Newbold-Davies experiment. 3. AR and MA processes and their properties. 4. ARIMA and SARIMA models: estimation, diagnostics and forecasting 5. ECM, VAR and VECM models: long-term relationships among time series, error correction model 6. volatility modeling: univariate GARCH models, diagnostics, extensions of GARCH models, practical applications (estimationg Value-at-Risk, option pricing) 7. volatility modeling: multivariate GARCH models (EWMA, DVEC, BEKK, CCC, DCC) 8. creating dynamic documents with Rmarkdown language |
Bibliography: |
1. Tsay (2013) An Introduction to Analysis of Financial Data with R 2. Biecek (2016) Przewodnik po pakiecie R 3. Tsay (2010) Analysis of Financial Times Series, Wiley 4. Brooks (2014) Introductory Econometrics for Finance, CUP 5. Gągolewski (2014) Programowanie w języku R 6. Suchwałko, Zagdański (2019) Analiza i prognozowanie szeregów czasowych |
Learning outcomes: |
After the course student will know: • what is stationarity of time series, white noise, autocorrelation and partial autocorrelation functions, • how ARIMA/SARIMA models are constructed • how ECM, VAR and VECM models are constructed • how GARCH models are constructed will understand: • conception of time-series cointegration and their long-term relationship • conception of the error correction mechanism • conception of conditional variance and univariate and multivariate models from GARCH family. will be able to: • estimate models from ARIMA/SARIMA famili, do diagnostic analysis and produce forecasts • assess ex-post quality of the forecast • estimate models from ECM/VAR/VECM family and interprete their results • estimate models from GARCH family, produce forecasts of conditional variance and apply the model to particular problems • create dynamic dokument with RMarkdown language. |
Assessment methods and assessment criteria: |
Home taken project and class activity |
Classes in period "Summer semester 2023/24" (in progress)
Time span: | 2024-02-19 - 2024-06-16 |
Navigate to timetable
MO TU W TH FR KON
KON
|
Type of class: |
Seminar, 30 hours
|
|
Coordinators: | Aneta Dzik-Walczak | |
Group instructors: | Aneta Dzik-Walczak | |
Students list: | (inaccessible to you) | |
Examination: |
Course -
Grading
Seminar - Grading |
Copyright by University of Warsaw.