Time series and forecasting**
A.A. 2025/2026
Obiettivi formativi
Forecasting time series data is of critical importance for a variety of decision-makers, and this course will focus on methodologies that can be applied to developing models for forecasting time series in a multitude of settings and applications.
Risultati apprendimento attesi
Upon completing this module, you will have the skills to:
1. Construct and validate both univariate and multivariate time series models.
2. Leverage time series models for forecasting future values.
3. Assess and compare forecasts generated by various models.
4. Generate point and density forecasts.
1. Construct and validate both univariate and multivariate time series models.
2. Leverage time series models for forecasting future values.
3. Assess and compare forecasts generated by various models.
4. Generate point and density forecasts.
Periodo: Primo quadrimestre
Modalità di valutazione: Esame
Giudizio di valutazione: voto verbalizzato in trentesimi
Corso singolo
Questo insegnamento non può essere seguito come corso singolo. Puoi trovare gli insegnamenti disponibili consultando il catalogo corsi singoli.
Programma e organizzazione didattica
Edizione unica
Responsabile
Periodo
Primo quadrimestre
Programma
Basics of forecasting and forecast evaluation
Forecasting with deterministic variables: trends and dummy variables
Forecasting with seasonal and non-seasonal ARMA models
Forecasting trends
Forecasting with dependent variables
Nonlinear models
Multivariate forecasting methods
Forecasting with deterministic variables: trends and dummy variables
Forecasting with seasonal and non-seasonal ARMA models
Forecasting trends
Forecasting with dependent variables
Nonlinear models
Multivariate forecasting methods
Prerequisiti
Although formal prerequisites are not necessary for this course, a fundamental understanding of matrix algebra, statistical inference, and econometrics will significantly improve your learning experience. In case you feel the need to review these topics, I suggest consulting the following textbook:
Metodi didattici
Lectures and tutorials using MATLAB.
Materiale di riferimento
Textbook: Hyndman, R. J., & Athanasopoulos, G. (2018). Forecasting: principles and practice. OTexts. Available online at: https://otexts.com/fpp3/
Matlab: https://work.unimi.it/servizi/servizi_tec/79539.htm
Supplementary materials might be provided during the course on the ARIEL platform
Matlab: https://work.unimi.it/servizi/servizi_tec/79539.htm
Supplementary materials might be provided during the course on the ARIEL platform
Modalità di verifica dell’apprendimento e criteri di valutazione
Written Exam
(An exception is made for the first session.)
First Exam Session - Alternative Evaluation Path
Only for the first session, students may choose an alternative to the standard written exam. This includes:
Project Paper: developed in groups of 2,
Oral Presentation: based on the project paper,
Individual Assignments.
Grading Breakdown (First Session Only):
Project Paper: 70%
Oral Presentation: 20%
Assignments: 10%
The project paper must focus on a time series forecasting problem.
The project topic will be defined during Week 3 of the course.
The oral presentation will take place at the end of the course. Each student must present individually in class and be prepared to answer questions about the project's content and results.
Important: After the first exam session, this alternative path will no longer be available. Students will need to take a written exam for all subsequent sessions.
Assignment Submission:
Assignments will be delivered through the ARIEL page of the course.
(An exception is made for the first session.)
First Exam Session - Alternative Evaluation Path
Only for the first session, students may choose an alternative to the standard written exam. This includes:
Project Paper: developed in groups of 2,
Oral Presentation: based on the project paper,
Individual Assignments.
Grading Breakdown (First Session Only):
Project Paper: 70%
Oral Presentation: 20%
Assignments: 10%
The project paper must focus on a time series forecasting problem.
The project topic will be defined during Week 3 of the course.
The oral presentation will take place at the end of the course. Each student must present individually in class and be prepared to answer questions about the project's content and results.
Important: After the first exam session, this alternative path will no longer be available. Students will need to take a written exam for all subsequent sessions.
Assignment Submission:
Assignments will be delivered through the ARIEL page of the course.
Docente/i