Dr. Gabriel B. GuillenEngineer · Attorney · Data Scientist

Academic Work

Research & Publications

Doctoral dissertation, Harvard capstone, and research papers spanning quantitative finance, epidemiology, social media analytics, and natural language processing.

Doctoral Dissertation

PhD Dissertation2021

Use of Bio-Inspired Metaheuristic Methods for Modeling and Prediction of High-Volatility Stock Markets: The Argentinian Case

Universidad Nacional de la Matanza — PhD in Economic Sciences

Doctoral thesis applying evolutionary algorithms, swarm intelligence, and other bio-inspired metaheuristic techniques to model and forecast price behavior in volatile equity markets, with a focus on the Argentine stock market (Merval index). Investigated the effectiveness of genetic algorithms, particle swarm optimization, and ant colony optimization for financial time-series prediction.

Research Papers & Projects

Capstone2022

Exploration of Lyme Disease Incidence Rate Modeling and Risk Assessment Mapping

Harvard University — Data Science Capstone

Dr. Gabriel Guillen, Yang Ming (Jason) Lin, Peter Masters

Developed predictive models for Lyme disease incidence using geospatial analysis, epidemiological data, and machine learning to produce county-level risk assessment maps across the Northeastern United States.

Research2021

Predicting COVID-19 Disease Spread Using Google Mobility Data and Clustering in Census Data

Harvard University — Data Science Research

Dr. Gabriel Guillen et al.

Applied clustering algorithms to Google Mobility Reports combined with U.S. Census demographic data to build predictive models for COVID-19 transmission rates at county and state levels during the pandemic.

Research2021

Benford's Law Applied to Twitter Data

Harvard University — Data Science Research

Dr. Gabriel Guillen et al.

Investigated whether naturally occurring Twitter engagement metrics (retweets, likes, follower counts) conform to Benford's Law, exploring applications for fraud detection and anomaly identification in social media data.

Research2021

Predicting U.S. Future Derivatives Market of Wheat Using NLP in President Trump Tweets

Harvard University — Data Science Research

Dr. Gabriel Guillen et al.

Used natural language processing (NLP) and sentiment analysis on the former U.S. President's social media output to build predictive models for wheat futures price movements, demonstrating the market impact of political communication.

Competition Paper2019

How to Predict Spread Movements

Lehigh University — 2019 IAFC Competition

Dr. Gabriel Guillen

Analyzed quantitative models for predicting bid-ask spread movements in equity and derivatives markets, presented at the International Association of Financial Colleges competition.

Research Methods & Tools

PythonRSQLJupyter NotebooksMachine LearningNLPTime-Series AnalysisGeospatial AnalysisClustering AlgorithmsGenetic AlgorithmsParticle Swarm OptimizationSentiment AnalysisFinancial EconometricsDerivatives ModelingBloomberg TerminalBig Data (Hadoop/Spark)Data Visualization