Open Science & Code Repositories
I am committed to open science and computational reproducibility. Below are public repositories containing code, replication packages, and dataset pipelines for my peer-reviewed publications and research projects.
NLP & Text Analytics
Text-Based ESG Narrative Indicators
Proposes an Environmental, Social, and Governance (ESG) indicator designed to overcome the limitations of traditional firm-specific ratings, such as inconsistency, infrequency, and bias toward large corporations. The study utilizes a scraped dataset over 390,000 online news articles covering the period from October 2004 to March 2024.
Mapping Central Ideas in Textual Corpora
Introduces a novel, field-agnostic tokenbased dictionary construction framework. The proposed pipeline integrates Part-of-Speech (POS) filtering, Term Frequency-Inverse Document Frequency (TF-IDF) weighting, and multi-word ngram extraction to systematically isolate core theoretical constructs without requiring exhaustive domain-specific neural network fine-tuning.
Internal consistency and stylistic stability of educational textbooks
Introduces a novel quantitative Textual Consistency Index based on a multi-lingual corpus of 480 historical economics textbooks across English, French, and Italian published between 1890 and 1976t. Utilizing a dual-pipeline natural language processing framework, we systematically quantify micro-structural coherence and pedagogical stability over nearly a century of economic thought.
Time Series & Econometrics
Data Pipelines & Web Scraping
Fetch Historical Data Using Python
This script is used to fetch historical data for various financial assets using the yfinance library.
Fetch Historical Data Using R
This script is used to fetch historical data for various financial assets using the quantmod library.
