Javier García-Bernardo
University of Amsterdam
Juyl 12th, 2019
computational social science
The CORPNET research group uncovers, investigates and aims to understand global networks of corporate power in contemporary global capitalism
What are we doing?
How?
- Using "big data" and bringing new methods from complex systems
- For the analysis of corporate ownership and corporate elites
Data collection
- Macro-data: e.g. Country indicators
- Micro-data: Surveys
- Linear regression
- By hand in Excel
"standard" social science framework
Data wrangling
Data
analysis
Theoretical understanding --> Hypothesis --> Test
(2) Modeling data-missingness
(6) Non-linear modeling
(4) Machine learning:
- Database merging
- Data classification
computational social science framework
Data collection
(1) New methods to collect data: Data scraping
Data wrangling
Data
analysis
(3) Set up appropriate data structures
(7) Motif-detection
(5) Description (data viz)
Data --> Patterns/Hypothesis --> (Generative mechanisms) --> Theory embedding
(1) New methods to collect data: Data scraping
(2) Modeling data-missingness
(3) Set up appropriate data structures
(4) Machine learning:
- Database merging
- Data classification
(5) Description (data viz)
(6) Motif-detection
(7) Non-linear modeling
- Ising model to understand information spread across the network of interlocks (w/ Rick Quax)
- Higher-order markov chains to understand combinations of countries that are important
- SBM to predict the tax rate using corporate networks (w/ Leto Peel)
...however this is our weaker part. Most of our research is descriptive.
https://CSSamsterdam.github.io/
