Panel data modelling provides a means to estimate and test relations when there are samples containing a time series of observations for a cross section of individuals, firms, markets, countries, etc. Current treatment for a selection of major topics will be presented, including endogeneity, serial correlation, stationarity and integration in addition to more traditional panel data techniques.
This training workshop is designed for practitioners and policy makers who are interested in gaining an in-depth knowledge of Panel Data techniques and learning how to use them in their current or future assignments.
At the end of the workshop, participants will be able to:
- Specify and estimate panel data models
- Perform relevant diagnostic tests for panel data models
- Use panel data models for policy decisions
- Use e-views and stata for panel models estimation.
- Introduction to panel data models
- Introduction to e-views and stata
- Traditional panel data model and diagnostic testing
- Pooled, fixed and random effects models
- Choosing between the fixed and random effects: the hausman test and the breusch and pagan (LM) test
- Dynamic panel data models and diagnostic testing
- Difference GMM estimator (Arellano and bond)
- System GMM estimator (Blundell and bond)
- Test of hypotheses and diagnostics: Hansen test and Sargan tests
- Non stationary panel data
- Panel unit root tests and co-integration
- Panel VAR, VECM models and causality
Methodology: Lectures and computer practice. Audio-visual aids will be used to reinforce these learning methods.
Econometricians, research officers, economists, academics, planning officers and policy analysts in both the public and private sectors.
Mode of Assessment for Certification: 100 per cent attendance and active participation.
Contact Person: T. A. Osanyintuyi and B. O. Olaniyan