 | | |
| MATH0221-3 | Analysis of time series
|
|
| Duration : | 30h Th, 10h Pr, 20h Mon. WS |
|
| Number of credits : |
|
|
| Lecturer : | Paul Gérard |
|
Language(s) of instruction :
|
| French language |
|
Course contents :
|
| 1. Modelisation using regression methods: decomposition into trend, seasonal component and perturbation. Detecting outlyers, forcasting.
2. Moving averages: properties and construction. Usual moving averages: arithmetic, Henderdon, Spencer,...
3. Exponential smoothing, Brown's methods, Holt&Winters methods.
4.Theory of stationary processes: stationarity, autocorrelation, partial autocorrelation, spectral analysis, ARMA, ARIMA and SARIMA processes. |
|
Learning outcomes of the course :
|
| Learning to analyse a time series using exploratory and modelling methods. Estimating its trend and seasonal components. Building and fitting ARIMA models. Making prediction. Estimating and using the power spectrum. |
|
Prerequisites and co-requisites/ Recommended optional programme components :
|
| A course on statistical inference and linear models.
For example: math0210-1 and math 213-1 |
|
Planned learning activities and teaching methods :
|
| Analysis time series with the methods presented during the course.
Introduction to the use of a statistical software on time series (Statistica). |
|
Mode of delivery (face-to-face ; distance-learning) :
|
| Lectures and homeworks.
Practical examples are treated during the lectures. |
|
Recommended or required readings :
|
| Handout based on transparencies.
Suggested reading:
1.Séries temporelles et modèles dynamiques. Christian Gourieroux et Alain Monfort. Economica, Collection " Economie et statistiques avancées ".
2.Time series : a biostatistical introduction. P.J.Diggle. Oxford statistical science series. Oxford University Press.
3.Practical time series. Gareth Janack. Arnold. |
|
Assessment methods and criteria :
|
| Oral evaluation and homeworks |
|
Contacts :
|
| Paul GERARD, Institut de Mathématique, Bât.B37, Grande Traverse 12, 4000-Liège
(Sart Tilman),Tel. 00-32-(0)4366.93.84, Fax : 00-32-(0)4366.95.47,
e-mail : paul.gerard@ulg.ac.be |
|
