Data-driven graduate admission tools. Replacing consultants with algorithms.
MFE admission prediction engine — GPBoost model (AUC 0.723) on 13,100+ records across 15 focused programs.
| Capability | Description |
|---|---|
| Admission Prediction | GPBoost model with per-program random intercepts and bias correction |
| Profile Scoring | 5-dimension evaluation across 37 sub-factors |
| School Ranking | Reach/target/safety with P(admit) confidence intervals |
| Prerequisite Matching | Course-by-course comparison against program requirements |
git clone https://github.com/MasterAgentAI/QuantPath.git
cd QuantPath && pip install -e .
quantpath predict # interactive mode — answer 8 questions, get resultsPython · GPBoost · LightGBM · scikit-learn · 465 tests · GitHub Actions CI
Built by Yicheng Yang — UIUC CS+Econ+Stats, Quantitative Researcher at Square Kettle LLC.
