The Econometric Analysis of Panel Data short course is of an applied nature and focuses on hands-on experience in estimation, interpretation and evaluation of economic relationships within a panel data context. The course covers techniques applicable to both stationary and non-stationary panel data sets, as well as static and dynamic model specifications; accounting for intercept and slope heterogeneity and cross-sectional dependence in the data and model specification.

We begin the discussion with the static linear model, where we consider both one-way and two-way error component models, where individual effects may be fixed or considered random factors. We further consider endogenous regressors, dynamic linear model relationships and sources of persistence; Nickel bias and correction; standard instrumental variables (IV) and general method of moments (GMM) estimation. We further consider models that extend heterogeneity in intercept coefficients to heterogeneity in slope coefficients, while also addressing cross-sectional dependence.
We conclude the course with a discussion of non-stationary in panels, focusing on panel unit roots assuming cross-section independence as well as cross-sectional dependence. We discuss the concept of spurious regression in panel data and panel cointegration tests. We conclude this section with a discussion on estimation and inference in panel cointegration models.
The course takes place in a computer lab on the main campus of the University of Pretoria. Delegates use EViews version 10 and Stata version 15 for practical applications.

After successful completion of this programme, you will be able to
• estimate regression models for data organised in a panel
• deal with violations of the basic assumptions of regression analysis, and
• apply techniques of unit root testing and cointegration.

The course covers the following topics:
1. Stationary panel data:
     • One-way error component models
     • Two-way error component models
     • Hypothesis testing

2. IV and dynamic panel data
     • Instrumental variables
     • Dynamic panel data models

3. Panel heterogeneity revisited, spatial dependence
     • Heterogeneity in slope coefficients
     • Cross-sectional dependence

4. Non-stationary panel data
     • Overview of the issues
     • Unit root tests
     • Estimation of non-stationary time series
     • Cointegration tests