79190 - Time Series

Learning outcomes

By the end of the course the student should know the fundamental theory of time series analysis. In particular the student should be able: - to analyse a time series in the time and in the frequency domain - to identify the stochastic process that has generated a time series based on the autocorrelation structure - to estimate and make inference on the parameters of a linear model for a stationary time series - to estimate time series components such as trend and seasonality by means of non parametric and parametric methods - to recognise the most important models for time series data

Course contents

The course covers the following topics. Linear models for time series data: linear processes, autoregressive unintegrated moving average processes (ARIMA) , seasonal processes. Identification, estimation and forecasting from ARIMA models. Time series decomposition. Time and frequency domain analysis.

 

Readings/Bibliography

Textbooks:

Brockwell P.J. and Davis R.A. (2002), Introduction to Time Series and Forecasting, Springer

Further readings:

Brockwell P.J. and Davis R.A. (1991). Time Series: Theory and Methods. Springer

 

 

Teaching methods

Lectures in person, online lectures, exercises, laboratory.

Assessment methods

Written exam, 2 hour-length, closed-books. 

Students may choose to present a group project that will be assessed. In this case, the final grade will be the weighted average of the grade in the group project (30%) and of the written exam (70%). 

The exam will be a pass if the final grade is greater or equal than 18.

Range of grades:

18-21 sufficient knowledge of the topics of the course 

21-24 discrete knowledge of the topics of the course

24-27 good knowledge of the topics of the course

27-30 excellent knowledge of the topics of the course

Teaching tools

Textbook, lecture notes, slides and auxiliary materials that can be found on the institutional teacher web-site and in Virtuale.

Office hours

See the website of Alessandra Luati