Introduction to Time Series Regression

MA Social Research Module 


Available to graduate students who have taken Data Analysis or equivalent.

Module Outline

This module introduces students to the analysis of time-series data using graphical and statistical techniques for model-fitting (regression).  The emphasis is on the practical application of these research techniques while striking a balance between intuition, statistical rigour and practical use of statistical software.  The module includes datasets that can be used to practise the statistical techniques taught on the course.  The module material also includes all the commands needed to implement these statistical techniques using EViews, Gretl, R-project, SPSS and Stata.

Learning Outcomes

On completion of the module, students should be able to:

  • Understand the special features of time-series models in terms of the dynamic structures that may become apparent;
  • Independently analyse time-series data using statistical software;
  • Understand the key features that make time-series regression different from cross-section regression and the possibility that the errors/residuals from the regression may be correlated across time;
  • Distinguish between time-series data that are stationary (mean-reverting to an average value) and time-series data that are non-stationary (increasing or decreasing perpetually in time);
  • Independently apply the basic techniques for including lagged values of the dependent and independent variables in time-series regression models;
  • Apply the basic tests used to identify the presence of autocorrelation in the errors/residuals of a time-series regression;
  • Understand the basic models used to accommodate the presence of autocorrelation in the errors/residuals in a time-series regression.

Course Assessment

Technical report involving time-series regression and not exceeding 2500 words including tables and equations.