Forecasting Methods

Objectives

The main objective of this course is to develop the skills needed to do empirical research in fields operating with time series data sets. The course intends to meet two goals. It provides tools for empirical work with time series data and is an introduction into the theoretical foundation of time series models. Much of statistical methodology is concerned with models in which the observations are assumed to be independent. However, many data sets occur in the form of time series where observations are dependent. In this course, we will concentrate on time series analysis, with a balance between theory and applications. After completing this course, a student will be able to analyze time series data using available software. In order to emphasize application of theory to real (or simulated) data, we will use R software.

General characterization

Code

100086

Credits

6.0

Responsible teacher

Bruno Miguel Pinto Damásio

Hours

Weekly - Available soon

Total - Available soon

Teaching language

Portuguese. If there are Erasmus students, classes will be taught in English

Prerequisites

Statistics and linear algebra (recomended)

Bibliography

Shumway, R.H. and Stoffer, D.S. Time Series Analysis and its Application with R Examples, 3rd edition, Springer, 2011. (http://www.stat.pitt.edu/stoffer/tsa3/); Makridakis, S., Wheelwright, S.C., Hyndman, R.J. Forecasting: Methods and Applications, 3rd edition, John Wiley & Sons, 1998.; Forecasting: principles and practice: https://www.otexts.org/book/fpp; Little Book of R for Time Series: http://a-little-book-of-r-for-time-series.readthedocs.org/en/latest/; Murteira, B., Muller, D., Turkman F. Análise de Sucessões Cronológicas, 1ª edição, McGraw Hill, 1993

Teaching method

The couse is based upon lectures and lab classes

Evaluation method

Continuous assessment:

(70%) 3 tests
(30%) 3 homework assignments

Avaliação final

(90%) Final exam (2nd round)
(10%) 3 homework assignments

Subject matter

1.    Time series basics: overview, autocorrelation and AR(1) model
2.    R tutorial
3.    Moving Average (MA) models and partial autocorrelation
4.    ARIMA Models: non-seasonal ARIMA models; diagnostics; forecasting
5.    Seasonal ARIMA models; identification
6.    Decomposition models
7.    Exponential smoothing
8.    The Periodogram
9.    Regression with ARIMA errors
10.    Two time series and cross-correlation
11.    Var models
12.    ARCH and GARCH models
13.    Longitudinal analysis
14.    Intervention analysis

Programs

Programs where the course is taught: