Time Series Analysis and Prediction

Recommended Prerequisites

Linear algebra, probability theory, Signal Processing, MATLAB

Teaching Methods

The teaching/learning process follows a learning approach of collaborative type, based on the presentation, analysis and discussion of concepts and techniques.
Theoretical classes (2 hours/week): Presentation of the concepts, principles and fundamental techniques for time series analysis and prediction. Examples of concrete situations will be presented, to illustrate the practical interest of the techniques and its application to real conditions.
Practical classes (2 hours/week): Practical problems addressing the theoretical concepts, their analysis, discussion and implementation.

Learning Outcomes

This course introduces fundamental concepts related to the theory, the  design and the implementation of time series analysis and prediction in various domains. The course covers linear and nonlinear methods for time series analysis, capable to build predictive models, in order to make inferences with application in prognosis and decision support systems.
This course will enable students to understand, identify, select and implement distinct time series and prediction methods and to apply these to practical situations in distinct domains.
The course will contribute to the acquisition of the following competences: 1. Instrumental: analysis and synthesis of complex problems. problem solving, namely in the prediction area; 2. Personal: team work, critical reasoning; 3. Systematic: self-learning, research.

Work Placement(s)

No

Syllabus

Introduction and concepts
Definitions, challenges and applications of time series analysis and prediction
Time series approaches, prediction methods

1| Linear Methods
Regressions, parameters, least squares estimation, correlations and autocorrelation
Order selection (information-theoretic and minimum description length criterions)
Stationarity and non-stationarity, trends
AR and ARMAX models
ARIMA models
Likelihood and confidence intervals
Univariate and multivariate dynamic regression models

2| Non-linear methods
Seasonality and exponential smoothing
Holt-winters method
Generalized autoregressive conditional heteroskedastic processes
Nonlinear dynamics and chaos
Non-linear regressive models - neural networks
Case based reasoning approaches

3| Comparison and Performance of prediction methods
Forecasting metrics, comparision of models, complexity measures

4| Applications
Prognosis for biomedical applications, economical predictions, finance, etc.

Assessment Methods

Assessment
Project: 50.0%
Exam: 50.0%

Bibliography

Linear Models and Time-Series Analysis Regression, ANOVA, ARMA and GARCH; Marc S. Paolella; 2019 JohnWiley & Sons Ltd; ISBN 9781119431985 (ePub) | ISBN 9781119431909

Time-Series Prediction and Applications - A Machine Intelligence Approach; Amit Konar, Diptendu Bhattacharya; 2017 Springer; ISBN 978-3-319-54596-7

Elements of Nonlinear Time Series Analysis and Forecasting; J. De Gooijer; 2017 Springer Series in Statistics; ISBN 978-3-319-43251-9

Nonlinear time Series analysis; Ruey S. Tsay, Rong Chen; 2019 John Wiley & Sons, Inc.; ISBN 9781119264064

Applied Time Series Analysis A Practical Guide to Modeling and Forecasting; Terence C. Mills; 2019 Elsevier Inc; ISBN: 978-0-12-813117-6