2023
Time Series
Name: Time Series
Code: MAT10174M
6 ECTS
Duration: 15 weeks/156 hours
Scientific Area:
Mathematics
Teaching languages: Portuguese
Languages of tutoring support: Portuguese, English
Regime de Frequência: Presencial
Sustainable Development Goals
Learning Goals
The aim of this course is to provide a varied set of statistical techniques so that he can identify various temporal patterns of a given phenomenon that evolves over time and, depending on this identification and the objectives of the analysis, apply the most appropriate stochastic model (s) to the data characteristics and study objectives.
Acquisition of fundamental theoretical concepts of stochastic processes and their importance in time series analysis. Study of mathematical models appropriate for modelling time series, univariate or multivariate, linear and nonlinear as well as study models for phenomena that evolve over time, considering the dependence between repeated observations intra individuals.
You should be able to use the appropriate software correctly. He / she should have acquired the competence to conduct autonomous study of other models, appropriate for solving practical cases that students may face in the future.
Acquisition of fundamental theoretical concepts of stochastic processes and their importance in time series analysis. Study of mathematical models appropriate for modelling time series, univariate or multivariate, linear and nonlinear as well as study models for phenomena that evolve over time, considering the dependence between repeated observations intra individuals.
You should be able to use the appropriate software correctly. He / she should have acquired the competence to conduct autonomous study of other models, appropriate for solving practical cases that students may face in the future.
Contents
1. Brief review of the essential concepts of Stochastic processes. Identification of temporal patterns. Temporal Decomposition Models
2. Linear models: ARMA, ARIMA and SARIMA
3. Non linear models: ARCH and GARCH
4. Temporal regression models
5. Analysis and modelling of time series, using the above methods, using the software R.
2. Linear models: ARMA, ARIMA and SARIMA
3. Non linear models: ARCH and GARCH
4. Temporal regression models
5. Analysis and modelling of time series, using the above methods, using the software R.
Teaching Methods
Lectures and practical classes taught in the blackboard and/or using slide show, accompanied by analysis and modelling of time series in R or other software that may prove to be more appropriate.
Introduction of theoretical concepts and application exercises using examples in various areas, thus seeking to sensitize students to the importance of the subject matter.
Introduction of theoretical concepts and application exercises using examples in various areas, thus seeking to sensitize students to the importance of the subject matter.
Assessment
Continuous evaluation: 3 individual each with their respective weights (20%, 40%, 40%).
Evaluation under examination: a final exam (30%) and one individual homeworks (70%).
Evaluation under examination: a final exam (30%) and one individual homeworks (70%).
Teaching Staff
- Dulce Maria de Oliveira Gomes [responsible]
