MS Economics · Northeastern University

Yun
Deng

Applied economist working at the intersection of causal inference, machine learning, and labor economics. First-year MS student at Northeastern University.

Scroll

About

Applied Economist
& Data Scientist

I am a first-year MS Economics student at Northeastern University (2025–2027), focused on applied microeconomics, labor economics, and the empirical methods that bridge economic theory with real-world data.

My research interests center on causal inference — specifically how machine learning methods like Double Machine Learning can sharpen causal estimates in observational settings. I enjoy working on projects that combine rigorous econometric frameworks with modern ML toolkits.

Outside of research, I am interested in health economics, labor market policy, and the economic implications of AI-driven occupational change.

4
Research Projects
3+
ML Frameworks
2027
Expected Graduation
NEU
Northeastern Univ.

Selected Projects

Research & Work

SELF-EMP.

Does Higher Occupational AI Exposure Cause Higher Probability of Self-Employment?

Causal inference analysis using Double Machine Learning (DML) and difference-in-differences (DID) to estimate the effect of AI occupational exposure on self-employment probability.

Causal InferenceDouble MLLabor EconomicsPython
POLICYNJPA

Minimum Wages and Employment

Card & Krueger Replication

Replication and extension of Card & Krueger (1994) using difference-in-differences. Explores heterogeneous treatment effects by restaurant ownership structure.

DiDLabor Economicsstatsmodels
XGBoostRFLRFALSE POSITIVE RATETPR
03

Depression Risk Prediction

NHANES 2015–2023

Predictive modeling of depression risk using 8 years of NHANES data. Compares Logistic Regression, Random Forest, and XGBoost with SHAP-based feature interpretation.

XGBoostRandom ForestSHAP
LOG SQUARE FOOTAGELOG PRICE
04

House Price Prediction

AHS 2023

Hedonic pricing models using AHS 2023 microdata. Benchmarks OLS against Ridge, Lasso, and Random Forest with log-space modeling to handle skewed price distributions.

OLSLasso/RidgeRandom Forestscikit-learn

Skills & Methods

Tools & Expertise

Programming

PythonRMATLABSQL

Econometrics & Causal Inference

Difference-in-DifferencesDouble Machine LearningInstrumental VariablesPanel Data MethodsRegression Discontinuity

Machine Learning

scikit-learnXGBoostSHAPRandom ForestRidge / Lasso

Data & Visualization

pandasNumPystatsmodelsmatplotlibStata

Contact

Let's Connect

I am open to research collaborations, internships, and conversations about applied economics. Feel free to reach out.