Panel econometrics is a very frequently used tool in economics, finance and political science. It allows us to analyze the statistical relationship between different variables on basis of a two-dimensional data set. Panel data comprise usually a time dimension (T) and a cross-sectional dimension (N). Thus, we observe over a defined period of time the characteristics of a given group of subjects (such as individuals, firms or countries).
This course aims to introduce quantitative methods and techniques used for panel estimations. The first part of the course starts with the simplest form of panel estimations, the so called `pooled ordinary least squares´ (pooled OLS) models. This part of the lecture gives the opportunity to review basic concepts of OLS regression analysis. The second part of this lecture focuses on traditional static linear panel models. We will learn about the estimation of fixed and random effects panel models and how to interpret their estimation results. In the third part of this course, we will discuss advanced panel econometrics comprising an introduction to dynamic panel models.
By the end of the course, students are expected to be able to explain theoretical concepts of panel modeling, to firmly implement panel estimations, and to interpret their estimation results. Accordingly, the course will put a strong emphasis on empirical applications. In the second part of each session, we will apply panel estimation techniques to different problem sets covering research questions in the field of macroeconomics, microeconomics and finance.
Students should have some basic understanding of econometric analysis (usually obtained by attending the course Statistics I and II). Knowledge of calculus, algebra and basic statistics are essential for this course. Experience with statistical software like STATA would be of great advantage, but is not essential.