B2163 - Econometrics of Panel Data

Learning outcomes

At the end of the course, students know the most appropriate estimating techniques for dynamic panel data models, both microeconomic (large and with more than one cross-sectional dimension) and macroeconomic (over a long time span). Specifically, they can: - critically understand theoretical and applied aspects of the vast literature based on dynamic panel data models; - apply dynamic panel data models techniques to their own analyses by programming specific routines using the STATA software.

Course contents

The course provides an overview of methodological and applied analyses on panel data, moving beyond standard FE and RE framework. Panels can be considered as clustered data, whether they are temporal observations nested within units or units nested within groups and supergroups. The key steps in guiding modeling strategies focus on decomposing total variability into variances between and within clusters, and assessing the presence of common factors and temporal dependence. Building on these foundations, the course covers:

variance decomposition and intraclass correlation;
new view on fixed-effects, random-effects and correlated random-effects models;
multilevel, hierarchical and cross-classified structures;
dynamic panel-data models;
heterogeneous slope models;
model comparison and specification strategies;
meta-analysis and meta-regression for synthesising empirical evidence and effect size;
reproducible empirical research using Stata/R.

The course combines methodological discussions with empirical applications based on real-world datasets drawn from economics, finance, innovation studies and environmental economics.

The course clearly requires the necessary prerequisities, especially for Erasmus students:

The course assumes previous knowledge of econometrics equivalent to introductory courses in microeconometrics, macroeconometrics and time-series econometrics. Students are expected to be familiar with topics including:
ordinary least squares;
hypothesis testing;
instrumental variables;
maximum likelihood estimation;
introductory panel-data methods;
basic time-series analysis.

Students enrolled in the Applied Economics and Markets programme are encouraged to review the contents of Econometrics for Individual Data and Time Series Econometrics. Students enrolled in the Economics and Econometrics programme are encouraged to review Econometric Methods, Microeconometrics and Macroeconometrics.

As a general reference, the minimum prerequisite knowledge is covered in Wooldridge J.M. 2020 Introductory Econometrics. A Modern Approach, Cengage, 7th Edition condenses the minimum requirement.

Basic familiarity with Stata is also expected. Useful references include: Baum, C. F. (2006) An Introduction to Modern Econometrics Using Stata, Stata Press; Cox N. J. (2001) Speaking Stata: How to repeat yourself without going mad, The Stata Journal, 1, Number 1, pp. 86–97.

Readings/Bibliography

Discussion papers, commented notes & slides, stata programmes and datasets will be available on the VIRTUALE platform and explained during the lectures.

Some references are

Snijders, T. and Bosker, R. (2012) Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Analysis, 2nd edition, Sage

Hansen, B. C. (2022) Econometrics, Princeton University Press, ch. 17.

Hanga, M. and Geyer-Klingeberga, J. and Rathgebera, A. W. and Stöcklb, S. (2018) Measurement matters—A meta-study of the determinants of corporate capital structure, The Quarterly Review of Economics and Finance, 68, 211-225

Yin, S. and Jia, F. and Chen, L. and Wang, Q. (2023) Circular economy practices and sustainable performance: A meta-analysis, Resources, Conservation and Recycling, 190, 106838

Teaching methods

Lectures combine theoretical discussion with hands-on emprirical applications unsing Stata/R.

Students are expected to actively participate in the implementation, interpretation and discussion of empirical analyses. Bringing a laptop to class is therefore strongly recommended.

The course adopts a research-oriented approach in which students progressively develop the methodological skills needed to design and conduct an empirical econometric study. 

Assessment methods

Assessment methods

Students may choose one of the following assessment pathways.

Continuous assessment, consisting of two components:

• Research assignment (30%)
• Individual oral presentation (70%)

Both components must be successfully completed by the January examination session immediately following the end of the course. After this session, the continuous assessment option is no longer available.

Standard examination

Students who do not complete the continuous assessment by the January examination session, or who choose not to participate in it, will be assessed through a written examination offered during the subsequent official examination sessions.

Research assignment (30%)

During the course, students will complete an empirical research project, either individually or in groups of up to three students. The research question may be proposed by the instructor or by the students, subject to the instructor's approval.

Students are required to justify their modelling choices, interpret their empirical results, and critically discuss the robustness and limitations of their analysis.

The use of Artificial Intelligence tools (e.g. ChatGPT, Claude, Gemini, Copilot, or similar systems) is permitted and encouraged as a support for learning and research. However, academic integrity requires full transparency. Students must therefore include, as an appendix to their report:

• the AI tool(s) used;
• the prompt(s) or prompting strategy employed;
• a description of how the AI-generated outputs were used;
• a critical evaluation of the accuracy, usefulness and limitations of the AI-generated suggestions;
• an explanation of which methodological and analytical decisions were made independently by the student(s).

The purpose of allowing AI tools is to develop students' ability to critically evaluate, verify and improve AI-generated suggestions through independent econometric reasoning.

Reproducible research: all empirical assignments must be fully reproducible. Students are expected to submit the Stata do-files used to generate all tables and results presented in their reports.

Individual oral presentation (70%)

The final assessment consists of an individual oral presentation (maximum 15 minutes) of the empirical panel-data analysis. The presentation assesses the student's ability to:

• formulate an appropriate econometric research question;
• select and justify suitable econometric methods;
• critically interpret empirical results;
• demonstrate an autonomous understanding of the underlying econometric techniques;
• discuss the strengths and limitations of the adopted methodology.

Particular emphasis is placed on the student's original reasoning, critical thinking, and independent mastery of the material. During the discussion, students may be asked additional questions concerning both the empirical application and the underlying econometric theory.

Standard written examination

The written examination assesses both theoretical knowledge and the ability to apply the econometric methods covered during the course to empirical problems. It is intended for students following the standard examination pathway.

Grading scale

• 30 cum laude: outstanding performance demonstrating complete mastery of the subject, original critical thinking, and excellent methodological understanding.
• 28–30: excellent performance showing comprehensive knowledge, sound methodological reasoning, and accurate interpretation of empirical evidence.
• 24–27: good performance demonstrating a solid understanding of the main concepts and methods.
• 18–23: satisfactory performance despite some theoretical or methodological weaknesses and limited critical discussion.
• Below 18: insufficient achievement of the learning outcomes

Teaching tools

Theoretical lectures are integrated with practical computer sessions during which students receive guidance in implementing econometric analyses in Stata.

Datasets, programming files and supplementary materials will be distributed through the University Virtuale platform.

A Microsoft teams virtual classroom will be available for exceptional cases in which students are unable to attend a lecture in person and for communication outside class.

Stata is available free of charge through the University licence: https://www.unibo.it/secure/software-stata/

Office hours

See the website of Maria Elena Bontempi