Time series and prediction with applications to epidemiology

Unit Coordinator: Umberto Triacca
ECTS Credits: 6
Semester: 1
Year: 2
Campus: University of L'Aquila
Language: English
Aims:

The aim of the course is to present important concepts of time series analysis (Stationarity of stochastic processes, ARMA models, forecasting etc.).

The course is a mixture of theory and practical applications of time series methods.

The theoretical part focuses upon properties of stationary time series and their analysis in the time domain. We use the econometric package GRETL.

At the end of the course, the student should be able to:

1. Compute and interpret a correlogramfor empirical applications.

2. Derive the properties of an ARIMA model

3. Select an appropriate ARIMA model for a given time series fit the model using an appropriate package

4. Compute forecasts

Content:

1. Stochastic processes (some basic concepts)

2. Stationary stochastic processes

3. Autocovariance and autocorrelation functions

4. Ergodicity of a stationary stochastic process

5. Estimation of moment functions of a stationary process

6. ARIMA models

7. Estimatiom of ARIMA models

8. Building ARIMA models

9. Forecasting from ARIMA models

Pre-requisites:

Prerequisite knowledge of basic inferential statistical methods.

Reading list:

The textbooks for the course are: 

  • Time Series Analysis Univariate and Multivariate Methods, 2nd Edition, W. Wei, 2006, Addison Wesley.
  • Time Series Analysis, J. Hamilton, 1994, Princeton University Press.
  • Time Series Analysis: Theory and Methods, P. Brockwell and R. Davis, 1991, Springer-Verlag.
  • Time Series Analysis and Its Applications with R Examples, Shumway, R. and Stofer, D., 2006, Springer.
  • Introduction to Time Series and Forecasting. Second Edition, P. Brockwell and R. Davis, 2002, Springer.

Print  

Related Articles