The Synthetic Gini Coefficient is an iniciative using Machine Learning techniques to estimate at the municipality level the Gini coefficient in Colombia. This is done for the purpose to contribute to persistent lack of inequality in Colombia. A working paper published in the following academic networks is developed to explain in detail the purpose and behavior of the synthetic estimation. Two machine learnings models are developed and one is Dominant in Fixed Effects while the other is Dominant in Varying Factors. The links of the municipality data sets are available here:
Munich RePEc Archive: https://mpra.ub.uni-muenchen.de/123561/
SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5128807
Cambridge Open Engage (With Datasets): https://www.cambridge.org/engage/coe/article-details/67a377d681d2151a02453d6d
Synthetic Gini Coefficient for the Dominant Fixed Effects model:
Synthetic Gini Coefficient under the Varying Factor model (Recommended for research):
A Looker Studio Dashboard has been developed to inspect the Varying Factor Model estimates of the Gini Coefficient. The link can be accessed next:
Riveros-Gavilanes, J. M. (2025). Municipality synthetic Gini index for Colombia: A machine learning approach. Munich Personal RePEc Archive, 1-42. URL: https://mpra.ub.uni-muenchen.de/123561/