Requirements
- Good laptop Attendance
Target audiences
- University students (BSc, MSc and PhDs) Research professionals Professionals in need of a refresher course on Applied Econometrics Professionals who want to enhance their Stata skills
- Instructor: admin
- Duration: 3 days
Panel data/longitudinal data refers to the data with observations about various cross-sections across time. One example is GDP data on East African Countries in the 1980-2020 period. For each cross-section (i.e. an entity: country, industry, or firm), information is collected for various time periods. Panel data has some advantages such as modeling both common and individual behavior of groups, having more information, more variability and more efficiency than pure time series or cross-sectional data, detecting and measuring statistical effects, and minimizing estimation biases.
This course will focus on the introduction to panel data with applications using the Stata software. Specifically, at the end of the course, participants are expected to have gained knowledge on the use of Stata to conduct the following: Panel data visualization, the distinction between time-variant and time-invariant variables, declaring data as panel dataset, graphical representation of heterogeneity, plotting, OLS regression, and its weakness, estimating and interpreting fixed effects and random effects, diagnostic tests, and tabulation
In this introductory course, only standard static panel data models will be covered. These, however, are not useful for example when we are dealing with dynamic panel data, when there is an endogeneity problem and when there is Nickel bias. In the second part, we shall explore models that address these issues
