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 the 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.

General characterization

Code

200088

Credits

3.5

Responsible teacher

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

Teaching method

The couse is based upon lectures and lab classes

Evaluation method

•    (60%) Final exam (1st or 2nd round dates)
•    (40%) Project

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