Machine Learning for Policy Research


As part of my coursework in machine learning at the University of Chicago, I explored diverse predictive models such as Linear Regression, Random Forest, and XGBoost to solve real-world policy problems. Projects included forecasting voter turnout, modeling COVID-19 impacts, and identifying tax evasion trends. This hands-on experience honed my skills in applying advanced algorithms, cross-validation, and hyperparameter tuning to generate actionable insights in public policy.

Topic Covered:
  1. Statistical Learning
  2. Linear Regression & Beyond Linearity
  3. Classification
  4. Resampling Methods
  5. Linear Model Selection & Regularization
  6. Tree-Based Methods
  7. Support Vector Machines
  8. Deep Learning

Tool Used:
  • Python (Scikit-learn, XGBoost, Pandas, NumPy, Matplotlib, Seaborn, Statsmodels, Joblib, Keras, imbalanced-learn)

Explore the projects on GitHub