English · 中文
Website ·
Research ·
Work ·
Practice ·
Contact
I am Qingjun, a student at Peking University preparing for doctoral-level research. My interests sit at the intersection of finance, statistical learning, and reproducible computation.
I am especially interested in questions where empirical design matters as much as model performance: how evidence is constructed, how assumptions shape conclusions, and how computational results can be made auditable.
My working principle: make the claim precise, make the evidence traceable, and make the result possible to challenge.
|
Financial machine learning, market microstructure, asset pricing signals, portfolio construction, and financial time series. |
Econometrics, statistical learning, interpretable models, robustness analysis, and uncertainty-aware evaluation. |
|
Reproducible experiments, provenance tracking, computational verification, and evidence-oriented workflows. |
Data leakage, temporal validation, transaction costs, multiple testing, and the gap between statistical and economic significance. |
An evidence-gated research skill for literature review, mathematical auditing, reproducible computation, and independent verification. It turns an open-ended investigation into a staged workflow with explicit claims, acceptance gates, and review checkpoints.
Python · Research methodology · Proof auditing · Reproducibility
Literature → Question → Formalization → Computation → Stress test → Communication
For substantial experiments, I aim to preserve the data source and version, preprocessing decisions, model configuration, random seeds, evaluation protocol, uncertainty, environment, and known failure cases.
My current toolkit includes Python, R, LaTeX, Git, Jupyter, Zotero, and the scientific Python ecosystem. Tools are secondary to the standard I want the work to meet: clear assumptions, honest limitations, and results that survive inspection.
- Strengthening foundations in probability, statistics, econometrics, optimization, and finance
- Reproducing empirical papers before extending their claims
- Developing research questions in quantitative finance and financial machine learning
- Exploring careful human–AI collaboration for literature discovery, verification, and research tooling
I welcome thoughtful conversations about quantitative finance, reproducible research, statistical learning, and research tools.
Email: phdstudytang@gmail.com
Website: studyer-tang.github.io
GitHub: @Studyer-Tang
Learn deeply · Build carefully · Verify everything