Statistics (Finance), Peking University

Homepage · Email · 中文

I study statistics and its applications to financial markets. My current interests include financial time series, market microstructure, and investment decisions. In empirical work, I pay particular attention to out-of-sample evaluation, data leakage, transaction costs, and reproducibility.

I also enjoy pure mathematics, particularly analysis, as well as history and deductive reasoning. I am especially fond of works by Bertrand Russell and Leonhard Euler.

• Statistical learning in finance: high-dimensional time series, factor models, and inference under non-stationarity.
• Market microstructure and financial behavior: order flow, liquidity, and adaptation among market participants.
• Sustainable investing and risk: ESG signals, asset dependence, and portfolio constraints.

• Advanced Mathematical Statistics Notes — Personal Chinese notes based on Fang Yao’s course lectures, with explanations, derivations, and examples. The first two chapters cover empirical distributions, statistical models, exponential families, sufficiency, and completeness. Includes PDFs, LaTeX sources, and a learning log; not official course notes or translations. Chapter 1 PDF · Chapter 2 PDF · Study progress.
• ThesisCraft — Local Word / WPS thesis formatting with institutional templates, typography, heading numbering, cross-references, and layout checks.
• PaperStage — A research presentation skill for AI assistants, with evidence-led narratives, academic layouts, and local editable PPTX tools.
• Academic Clipboard — A local-first clipboard for text, screenshots, and research snippets, including source annotations and DOI, BibTeX, equation, and table conversions.

• AutoElectiveOrb — A Windows course assistant with availability monitoring, lottery-result viewing, and pre-swap checks. CAPTCHA recognition uses a third-party online service.
• Codex Daily Token Widget — A Windows widget and web dashboard for daily, task, and turn-level token usage from local session logs; these counts are not API billing or subscription quota figures.
• DropOrb — A Windows drop orb for files, clipboard content, and quick actions, with local processing.

Project list reviewed on 2026-09-29. See each repository for current capabilities and limitations, or browse all repositories.

Selected reading in analysis, statistics and history, not a record of completed reading:

• Elias M. Stein & Rami Shakarchi, Real Analysis — measure, integration and Hilbert spaces; assumes undergraduate analysis.
• Haim Brezis, Functional Analysis, Sobolev Spaces and Partial Differential Equations — graduate-level analysis; assumes real analysis and Lebesgue integration.
• Larry Wasserman, All of Statistics — a concise overview of statistical inference; assumes calculus and linear algebra.
• Gareth James et al., An Introduction to Statistical Learning — applied statistical learning; free R and Python editions on the authors’ website.
• Marc Bloch, The Historian’s Craft — historical inquiry and source criticism, published by Manchester University Press.

For games and deduction: Return of the Obra Dinn, recommended for its reasoning mechanics, not as an academic or historical reference.

• Texts: Stephen Abbott, Understanding Analysis; Sheldon Axler, Measure, Integration & Real Analysis; Hastie, Tibshirani & Friedman, The Elements of Statistical Learning.
• Russell and Euler: Russell, Introduction to Mathematical Philosophy and The Problems of Philosophy; Euler, Introduction to Analysis of the Infinite and Elements of Algebra. Historical works to read alongside modern mathematics, not replacements for current textbooks.
• Novels: Jorge Luis Borges, Ficciones; Fyodor Dostoevsky, Crime and Punishment; Umberto Eco, The Name of the Rose.
• History: Natalie Zemon Davis, The Return of Martin Guerre (microhistory).

Prerequisites, descriptions and available source links are on my academic homepage. These are recommendations, not a completed reading list.