Open to opportunities

Meng Song

@mengsong

Message

Econometrics Ph.D. candidate applying causal inference and machine learning to inform policy and data-driven decisions.

United States

Message

What I'm looking for

I’m looking for full-time work where I can combine econometrics, causal inference, and ML/NLP to evaluate policies and build data-driven models that support real business and societal decisions.

I’m a Ph.D. candidate in Econometrics and Quantitative Economics with a Master’s degree in Data Science, specializing in empirical economics, causal inference, and policy evaluation. My work focuses on turning complex data into clear, decision-ready evidence.

At Payactiv, I built a FinTech risk analytics framework for earned wage access transactions, developing ML models in Python (including logistic regression, decision trees, and XGBoost) to support employee-level risk classification and loss prediction. I also used SQL to prepare payroll and disbursement data, engineer repayment-behavior features, and design/evaluate an approve-review-decline decision framework.

Through the UConn Institute for Collaboration on Health, Intervention, and Policy, I design and analyze RCTs and A/B tests using causal inference methods. I clean and integrate multi-source datasets with Stata, R, and SPSS, and I develop dashboards and research materials to communicate results on gun injury prevention.

My research includes large-scale causal methods (stacked DiD, 2WFE, regression discontinuity design, propensity score matching) and scalable geospatial pipelines—geocoding 1+ billion addresses using the Google Geocoding API and Pelias Geocoder with Python and MongoDB. I’ve authored research on work-from-home and labor-force participation and are actively developing projects using generative AI, LLM text analysis, and advanced ML approaches.

Experience

Work history, roles, and key accomplishments

Current

Quantitative Graduate Assistant

Current

UConn Institute for Collaboration on Health

May 2025 - Present (1 year 1 month)

Designed and analyzed randomized controlled trials and A/B tests using causal inference methods, cleaning and integrating multi-source datasets. Built codebooks, surveys, and interactive dashboards to communicate gun injury prevention findings.

Causal Inference A B Testing Data Cleaning Stata R SPSS Survey Design Dashboards

Current

Research Assistant

Current

University of Connecticut

May 2023 - Present (3 years 1 month)

Applied causal inference methods including stacked DiD, two-way fixed effects, regression discontinuity, propensity score matching, and principal component analysis. Built scalable Python geocoding and storage pipelines using Google Geocoding API, Pelias, and MongoDB for geospatial data processing.

Stacked DiD Two Way Fixed Effects Regression Discontinuity Principal Component Analysis geocoding MongoDB Geospatial Data Processing

Data Science Intern

Payactiv, Inc.

Jun 2026 - Aug 2026 (2 months)

Developed a FinTech risk analytics framework for earned wage access, using Python machine learning models to classify employee risk and predict transaction-level loss. Used SQL for payroll/disbursement feature engineering and designed an approve/review/decline decision framework balancing risk, approval rates, and business impact.

Machine Learning Logistic Regression Decision Trees XGBoost SQL Feature Engineering Risk & Analytics

Education

Degrees, certifications, and relevant coursework

University of Connecticut

Master of Science (M.S.), Data Science

2026 -

M.S. in Data Science with work focused on applied machine learning and data-driven research methods. Program graduation is flexible between Dec 2026 and May 2027.

University of Connecticut

Doctor of Philosophy (Ph.D.), Econometrics and Quantitative Economics

Ph.D. candidate in Econometrics and Quantitative Economics (STEM). Focuses on empirical economics, causal inference, policy evaluation, experiment design, and applied ML/NLP methods, with flexible graduation in Dec 2026 or May 2027.