https://github.com/chirindaopensource/measuring_corruption_from_text_data

End-to-End Python implementation of Muço’s (2025) corruption measurement framework. Combines NLP pipeline (regex extraction, Porter stemming, TF-IDF), PCA-based dimensionality reduction, and fixed-effects OLS to quantify institutional quality from Brazilian audit reports. Includes supervised learning robustness checks and LOO sensitivity analysis.

audit-analysis brazilian-data corruption-measurement dictionary-based-classification dimensionality-reduction econometrics fixed-effects government-transparency institutional-quality natural-language-processing nltk political-economy portuguese-nlp principal-component-analysis research-replication scikit-learn supervised-learning text-as-data text-classification text-mining

Last synced: 27 Apr 2026