Econometrics of Networks

Note: all our courses are taught online this year.

Networks play an increasingly dominant role in many social, business, and economic environments. Moreover, network data becomes increasingly important and available due to the rise of online social media and digitization. This course offers a concise introduction into the most recent econometric methods developed for processing, visualizing and learning from network data. The course will combine online lectures with hands-on empirical and programming exercises.

Course days11-15 January 2021
Course levelAdvanced undergraduate and graduate students, PhD candidates, data scientists and professionals from all disciplines 
Coordinating lecturerMichael König
Other lecturersTba
Forms of tuitionOnline lectures and integrated exercises
Forms of assessmentData and programming exercises
Credits2 ECTS
Contact hours20 hours
Tuition feeRead all information about our tuition fees and what's included here  
How to applyFind our application form here 
This course is suitable for graduate, as well as advanced undergraduate students, data scientists and practitioners from all areas who are interested in the analysis of networks. The course combines recent statistical and econometric methods with sound economic micro-foundations to analyze the role of networks in various social and economic environments. All parts of the course (economic theory and econometrics) will be presented in a consistent and interdependent framework, which will make clear that none of them should be studied in isolation. Students becoming more familiar with one particular area will benefit from insights gained through complementary views from other areas. The course will be accessible to advanced undergraduate and graduate students as well as practitioners from all related fields. If you have doubts about your eligibility for the course, please contact us: [email protected]

1. Examples of Networks and Data 

2. Network Statistics, Visualization and Graphs
• Elements of Graph Theory
• Graphs and Matrices 
• Bipartite Graphs 
• Core-periphery Networks and Nested Split Graphs 
• Network Statistics: Average path length, clustering and assortativity 
• Centrality in Networks: Degree, eigenvector, Katz-Bonacich centrality and Google's Page Rank 
• Network Visualization: Force-directed, circular and layered layout 

3. Econometrics of Interactions in Networks
• Spatial Autoregressive (SAR) Model  
• Linear Quadratic Utility  
• Endogeneity of the Spatial Lag  
• Two-Stage Least Squares (2SLS)  
• Maximum Likelihood Estimation (MLE) 
• Identification Issues   
  – Correlated Effects, Sorting and Selection     
  – Endogenous Link Formation  
• Multiple Spatial Weight Matrices  
• Spatial Panel Data 

4. Econometrics of Network Formation
• Exponential Random Graph Model (ERGM) 
• Conditional Edge-Independence     
   – Erdös-Rényi Random Graph     
   – Logistic Regression     
   – Unobservable Characteristics (beta-model) 
   – Tetrad Logit Estimator  
• Random Utility Model 
• Maximum Likelihood Estimation (MLE)  
• Markov Chain Monte Carlo: Gibbs Sampling and the Metropolis Hastings Algorithm 
• Stochastic Block Model (SBM)  
• Temporal ERGMs 

5. Joint Estimation of Outcomes and Network Formation
5.1 Coevolution of Networks and Behavior: An application to R&D collaboration networks 
• Structural Model: Utility and the potential game 
• Estimation   
   – Computational Problem and the Exchange Algorithm     
   – Double Metropolis-Hastings (DMH) Algorithm     
   – Unobserved Heterogeneity  
• Empirical Illustration: R&D collaborations     
5.2 Network Formation with Multiple Activities: An application to team production and co-authorship networks
• Bipartite Network, Production Function, and Utility
• Equilibrium Characterization and Line Graphs  
• Estimation with Endogenous Matching
• Empirical Illustration: Co-authorship networks

6. Big Data meets Networks
6.1 The Digital Layer: How innovative firms relate on the Web
6.2 Using Text-mining Algorithms to Identify R&D Collaborations from the News


Upon successful completion of the course, students will:

  • become acquainted with different statistical methodologies for analyzing networks while learning how to see these different methodologies complementing each other. 
  • learn to model network problem situations mathematically, and adapt the methods learned to new situations at hand. 
  • be able to recognize, understand, and analyze societal and business problems in which networks are central. 
  • learn how networks affect demand and supply in markets, how this leads to market failures, and how government policies can address these. 

All relevant material will be covered in the lecture slides. The slides will be made available to the students on the course website before the start of the course. The following literature is complementary to the course slides and covers some additional relevant material for further reading: 
  
• Graham, Bryan, and De Paula, Aureo. Econometric Analysis of Network Data. Elsevier, 2020. 
• Kolaczyk, Eric, Statistical Analysis of Network Data: Methods and Models, Springer, 2009.  
• LeSage, James, and Robert Kelley Pace. Introduction to Spatial Econometrics. Chapman and Hall/CRC, 2009. 
• Bramoulle, Yann, Andrea Galeotti, and Brian Rogers. The Oxford Handbook of the Economics of Networks. Oxford University Press, 2016.  
• Jackson, Matthew. O. Social and Economic Networks. Princeton University Press, 2010. 
• Carrington, Peter, John Scott, and Stanley Wasserman. Models and Methods in Social Network Analysis, Cambridge University Press, 2005. 

Michael D. König is Associate Professor at the Department of Spatial Economics at VU, Amsterdam. He is also a research fellow at the Tinbergen Institute, the Centre for Economic Policy Research (CEPR) in London, and the Swiss Economic Institute (KOF/ETH) in Zurich. Prior to joining the VU Amsterdam, he was a visiting scholar at the Stanford Institute for Economic Policy Research (SIEPR) and the Department of Economics at Stanford University. His research focuses on the economics of innovation and technical change, and how these affect and are being effected by networked relationships between various economic actors, ranging from individuals to firms, and from sectors to countries. His research combines both theoretical as well as empirical methods, and he uses these methods to evaluate real-world policy instruments. 

Konig