docs: make Chinese README the default

README.md

@@ -1,361 +1,263 @@
-# Machine Learning for Trading — 3rd Edition
+<!-- WEHUB_ZH_README -->
+> [!NOTE]
+> 本文档由 WeHub 基于上游 README 翻译整理，属于社区翻译，非官方中文文档。  
+> [English](./README.en.md) · [原始项目](https://github.com/stefan-jansen/machine-learning-for-trading) · [上游 README](https://github.com/stefan-jansen/machine-learning-for-trading/blob/HEAD/README.md)  
+> 原作者、版权与许可证归属以原始项目及本仓库 LICENSE 文件为准。
 
-**Build, test, and deploy ML-driven trading strategies — from data sourcing to live execution.**
+# Machine Learning for Trading — 第三版
 
-This repository hosts the code for [*Machine Learning for Trading, 3rd Edition*](https://amzn.to/4eigy2F)
-by [Stefan Jansen](https://www.linkedin.com/in/applied-ai/) — a ground-up
-rebuild, organized around one end-to-end workflow: how you define a research idea and develop it iteratively into a
-strategy you can actually run, and keep running, in a live market.
+**构建、测试并部署机器学习驱动的交易策略——从数据获取到实盘执行。**
 
-- [Nine case studies](https://www.ml4trading.io/case-studies/) illustrate the workflow throughout the 27 chapters of the
-  book, from raw data through features, models, backtests, costs, and risk to deployment.
-- **Generative AI** and **autonomous agents** are new to this edition and cut across that workflow, bringing
-  retrieval-augmented generation, knowledge graphs, and multi-agent systems to financial research.
-- The [companion website](https://ml4trading.io) features [112 primers](https://ml4trading.io/primer/),
-  [56 agent skills](https://ml4trading.io/skills/),
-  and [six production Python libraries](https://ml4trading.io/libraries/)
-  that facilitate substantial parts of the workflow.
+本仓库托管 [*Machine Learning for Trading, 3rd Edition*](https://amzn.to/4eigy2F) 一书的代码，作者为 [Stefan Jansen](https://www.linkedin.com/in/applied-ai/) ——这是一次从零开始的重构，围绕一条端到端工作流组织：如何将研究构想定义出来，并通过迭代开发成一套能在真实市场中实际运行、并持续运行的策略。
+
+- [九个案例研究](https://www.ml4trading.io/case-studies/) 贯穿全书 27 章展示该工作流，涵盖从原始数据、特征、模型、回测、成本与风险到部署的完整路径。
+- 本版新增的 **生成式 AI（Generative AI）** 与 **自主智能体（autonomous agents）** 贯穿该工作流，将检索增强生成（retrieval-augmented generation）、知识图谱与多智能体系统引入金融研究。
+- [配套网站](https://ml4trading.io) 提供 [112 篇入门讲解](https://ml4trading.io/primer/), [56 项 agent 技能](https://ml4trading.io/skills/), 以及 [六个可用于生产的 Python 库](https://ml4trading.io/libraries/)，可支撑工作流中的多个重要环节。
 
 <p align="center">
-  <a href="https://amzn.to/4eigy2F"><img src="assets/cover.png" width="45%" alt="Machine Learning for Trading, 3rd Edition"></a>
+  <a href="https://amzn.to/4eigy2F"><img src="assets/cover.png" width="45%" alt="Machine Learning for Trading, 第三版"></a>
 </p>
 
-## 🎓 New: Live Courses & Lightning Lessons
+## 🎓 新增：直播课程与闪电课
 
-For the first time, the third edition comes with a **live cohort course**, hands-on **workshops**, and free
-**lightning lessons** taught by Stefan on [Maven](https://maven.com/stefan-jansen) — full schedule on the
-[courses page](https://ml4trading.io/courses/).
+第三版首次配套 **直播同期班课程**、动手 **工作坊** 以及由 Stefan 在 [Maven](https://maven.com/stefan-jansen) 平台开设的免费 **闪电课** ——完整课表见 [课程页面](https://ml4trading.io/courses/).
 
-- **▶ [Machine Learning for Trading: From Research to Production](https://maven.com/stefan-jansen/research-to-production)**
-  — the flagship live cohort course: take a research idea all the way to a deployed, monitored strategy, working
-  through the book's end-to-end workflow with direct feedback. **The first cohort starts Monday, July 6, 2026 —
-  enrollment closes Friday, July 3.**
-- **[Getting Stuff Done with Coding Agents](https://maven.com/p/8394ac/getting-stuff-done-with-coding-agents?utm_medium=ll_share_link&utm_source=instructor)**
-  — a free lightning lesson on putting coding agents to work.
-- **[Building Multi-Agent Forecasting Systems](https://maven.com/stefan-jansen/forecasting-agents)**
-  — a hands-on workshop on engineering the forecasting-agent loop: building auditable, debate-driven multi-agent
-  systems for financial research.
+- **▶ [Machine Learning for Trading: From Research to Production](https://maven.com/stefan-jansen/research-to-production)** ——旗舰直播同期班课程：将研究构想一路推进到已部署、可监控的策略，在直接反馈下走完书中的端到端工作流。**首期班于 2026 年 7 月 6 日（周一）开课——报名于 7 月 3 日（周五）截止。**
+- **[Getting Stuff Done with Coding Agents](https://maven.com/p/8394ac/getting-stuff-done-with-coding-agents?utm_medium=ll_share_link&utm_source=instructor)** ——一节关于如何让 coding agents 投入实战的免费闪电课。
+- **[Building Multi-Agent Forecasting Systems](https://maven.com/stefan-jansen/forecasting-agents)** ——动手工作坊，讲授如何工程化预测智能体循环：为金融研究构建可审计、辩论驱动的多智能体系统。
 
 <p align="center">
-  <a href="https://youtu.be/Ksxv9QVZSOo"><img src="assets/course-trailer.jpg" width="60%" alt="Watch the course overview: Machine Learning for Trading — From Research to Production"></a>
+  <a href="https://youtu.be/Ksxv9QVZSOo"><img src="assets/course-trailer.jpg" width="60%" alt="观看课程概览：Machine Learning for Trading — From Research to Production"></a>
 </p>
 
 ---
 
-## What's New in the Third Edition
+## 第三版有哪些新内容
 
-The whole book traces one path: from data infrastructure and strategy research, across an *evidence boundary* that
-separates tuning from evaluation, to deployment and monitoring — with a feedback loop that retrains, pauses, or
-retires a strategy as its edge decays.
+全书遵循一条统一路径：从数据基础设施与策略研究出发，跨越将调优与评估分开的 *证据边界（evidence boundary）*，直至部署与监控——并辅以反馈循环，在策略优势衰减时进行再训练、暂停或退役。
 
 <p align="center">
-  <img src="assets/workflow.png" width="90%" alt="The ML4T workflow: data infrastructure and strategy research, an evidence boundary separating tuning from evaluation, and deployment with a retrain/pause/retire feedback loop">
+  <img src="assets/workflow.png" width="90%" alt="ML4T 工作流：数据基础设施与策略研究、将调优与评估分开的证据边界，以及带有再训练/暂停/退役反馈循环的部署">
 </p>
 
-Where earlier editions moved technique by technique, the third edition runs that one process end to end — and adds
-substantial new material:
+早期版本按技术逐点推进，第三版则将这一流程端到端贯通，并新增大量内容：
 
-- **A wider model toolkit**: from gradient boosting (XGBoost, LightGBM, CatBoost) to deep time-series architectures
-  (PatchTST, iTransformer, TSMixer, TCN, Mamba) and newer tabular and latent-factor models (TabPFN, TabM, conditional
-  and supervised autoencoders).
-- **Dedicated strategy-design chapters**: transaction costs and risk management are now full chapters, neither of
-  which existed before, joining portfolio construction and strategy synthesis so a raw signal is carried through to a
-  sized, cost- and risk-aware portfolio.
-- **A full production track**: live trading systems (Interactive Brokers, Alpaca, QuantConnect), MLOps and governance
-  (drift detection, safe rollout, circuit breakers, feature stores, experiment tracking), and the operational reality
-  of *running* strategies, not just building them.
-- **Generative AI**: retrieval-augmented generation grounded in SEC filings, knowledge graphs and Graph RAG, and
-  autonomous, multi-agent research systems.
-- **Causal machine learning**: Double ML, Bayesian structural time series, and causal discovery for separating real
-  effects from spurious correlation.
-- **Reinforcement learning**: optimal execution, market making with inventory, and deep hedging.
-- **Synthetic financial data**: TimeGAN, Tail-GAN, Sig-CWGAN, and diffusion-based generators for validation when
-  history is short.
+- **更丰富的模型工具箱**：从梯度提升（XGBoost、LightGBM、CatBoost）到深度时间序列架构（PatchTST、iTransformer、TSMixer、TCN、Mamba），再到较新的表格与潜因子模型（TabPFN、TabM、条件自编码器与监督自编码器）。
+- **专门的策略设计章节**：交易成本与风险管理现为完整章节（此前均未单独成章），并与组合构建、策略综合衔接，使原始信号一路贯通至经规模调整、成本与风险感知的组合。
+- **完整的生产化路径**：实盘交易系统（Interactive Brokers、Alpaca、QuantConnect）、MLOps 与治理（漂移检测、安全发布、熔断机制、特征存储、实验跟踪），以及*运行*策略（而非仅构建策略）的运维现实。
+- **生成式 AI**：基于 SEC 申报文件的检索增强生成、知识图谱与 Graph RAG，以及自主多智能体研究系统。
+- **因果机器学习（Causal machine learning）**：Double ML、贝叶斯结构时间序列（Bayesian structural time series）与因果发现，用于区分真实效应与虚假相关。
+- **强化学习（Reinforcement learning）**：最优执行、带库存的做市，以及深度对冲。
+- **合成金融数据**：TimeGAN、Tail-GAN、Sig-CWGAN 以及基于扩散的生成器，用于在历史数据较短时进行验证。
 
-Methodological rigor is treated as a first-class topic rather than an afterthought. The book draws an explicit line
-between exploration and confirmation — the *evidence boundary* — uses walk-forward cross-validation throughout, and
-confronts the multiple-testing and overfitting problems that quietly invalidate most backtests, with tools like the
-Deflated Sharpe Ratio, the Rademacher Anti-Serum, and White's Reality Check, plus conformal prediction for honest
-uncertainty estimates.
+方法论严谨性被作为一等公民主题，而非事后补充。本书明确区分探索与确认——即 *证据边界*——全书采用滚动前向交叉验证（walk-forward cross-validation），并直面悄然使大多数回测失效的多重检验与过拟合问题，借助 Deflated Sharpe Ratio、Rademacher Anti-Serum、White's Reality Check 等工具，以及保形预测（conformal prediction）以获得诚实的 uncertainty 估计。
 
-The data layer moves to **Polars** for fast, expression-based manipulation, and every chapter ships in **reproducible
-Docker environments** so results repeat across machines; PyTorch, LightGBM, Optuna, and Plotly round out the modeling
-and visualization stack.
+数据层迁移至 **Polars**，以实现快速的基于表达式的数据处理；每章均提供 **可复现的 Docker 环境**，确保结果在不同机器上可重复；PyTorch、LightGBM、Optuna 与 Plotly 共同构成建模与可视化技术栈。
 
-### Nine Case Studies
+### 九个案例研究
 
-The structural centerpiece of the third edition is **nine case studies** that run the length of the
-book. ETFs, crypto
-perpetuals, intraday equities, options, FX, futures, and equity factor panels are each carried through the *same*
-pipeline — from raw data and labels to features, models, backtests, costs, risk overlays, and a final deployment
-assessment. One disciplined process applied to nine very different markets shows where it works, where it breaks, and
-why.
+第三版的结构核心是 **九个案例研究**，贯穿全书。ETF、加密货币永续合约、日内股票、期权、外汇、期货与股票因子面板均经由*同一*流水线——从原始数据与标签到特征、模型、回测、成本、风险叠加，直至最终部署评估。将一套严谨流程应用于九个截然不同的市场，可展示其适用之处、失效之处及原因。
 
-| Case Study               | Asset Class        | Frequency | What It Explores                                                             |
+| 案例研究 | 资产类别 | 频率 | 探索内容 |
 |--------------------------|--------------------|-----------|------------------------------------------------------------------------------|
-| ETFs                     | Multi-asset ETFs   | Daily     | Cross-asset momentum and mean-reversion across 100 ETFs                      |
-| Crypto Perps             | Crypto             | 8-hourly  | Funding-rate arbitrage on perpetual futures                                  |
-| NASDAQ-100               | Equities           | 15-min    | Intraday microstructure signals from order flow and the LOB                  |
-| S&P 500 Equity + Options | Equities + Options | Daily     | Equity selection enhanced with implied-volatility features                   |
-| US Firm Characteristics  | Equities           | Monthly   | Firm-level characteristics panel (size, value, momentum, quality)            |
-| FX Pairs                 | FX                 | Daily     | Carry and momentum across major currency pairs                               |
-| CME Futures              | Futures            | Daily     | Term-structure and roll-yield signals across commodity and financial futures |
-| S&P 500 Options          | Options            | Daily     | Options-only strategies (straddles, delta-hedged positions)                  |
-| US Equities              | Equities           | Daily     | Broad cross-section of US stocks with classic factor exposures               |
+| ETFs | 多资产 ETF | 日频 | 100 只 ETF 的跨资产动量与均值回归 |
+| Crypto Perps | 加密货币 | 8 小时 | 永续期货的资金费率套利 |
+| NASDAQ-100 | 股票 | 15 分钟 | 来自订单流与 LOB 的日内微观结构信号 |
+| S&P 500 Equity + Options | 股票 + 期权 | 日频 | 以隐含波动率特征增强的股票筛选 |
+| US Firm Characteristics | 股票 | 月频 | 公司层面特征面板（规模、价值、动量、质量） |
+| FX Pairs | 外汇 | 日频 | 主要货币对的 carry 与动量 |
+| CME Futures | 期货 | 日频 | 商品与金融期货的期限结构与 roll-yield 信号 |
+| S&P 500 Options | 期权 | 日频 | 纯期权策略（跨式、delta 对冲头寸） |
+| US Equities | 股票 | 日频 | 美国股票广截面及经典因子暴露 |
 
-### 112 Primer Topics
+### 112 个入门主题
 
-Free concept explainers for every idea the book relies on. Each part links to its full list; a few topics show the
-range:
+为书中所依赖的每个概念提供免费讲解。各部分均链接至完整列表；以下若干主题展示其覆盖范围：
 
-- [Foundations](https://ml4trading.io/primer/): 8 topics spanning limit order book mechanics, bitemporal data models,
-  and the stylized facts a simulator must reproduce.
-- [Research Design and Feature Engineering](https://ml4trading.io/primer/): 21 topics, including multiple testing in
-  factor research, fractional differencing, and path signatures for financial sequences.
-- [Model Development](https://ml4trading.io/primer/): 22 topics, among them regularization geometry, conformal
-  prediction in finance, and the mechanism behind double machine learning.
-- [Strategy Implementation](https://ml4trading.io/primer/): 27 topics, from the deflated Sharpe ratio and hierarchical
-  risk parity to Almgren-Chriss optimal execution.
-- [Advanced AI](https://ml4trading.io/primer/): 8 topics such as Markov decision processes, the policy-gradient theorem,
-  and proper scoring rules for event forecasts.
-- [Production](https://ml4trading.io/primer/): 2 topics, champion-challenger evaluation and training-serving skew with
-  feature stores.
-- [Cross-cutting concepts](https://ml4trading.io/primer/): 20 building blocks referenced across chapters, for example
-  momentum and mean reversion, the bias-variance tradeoff, and walk-forward validation.
+- [基础](https://ml4trading.io/primer/): 8 个主题，涵盖限价订单簿机制、双时态数据模型，以及模拟器必须复现的 stylized facts。
+- [研究设计与特征工程](https://ml4trading.io/primer/): 21 个主题，包括因子研究中的多重检验、分数差分，以及金融序列的路径签名（path signatures）。
+- [模型开发](https://ml4trading.io/primer/): 22 个主题，其中有正则化几何、金融中的保形预测，以及双重机器学习（double machine learning）的机制。
+- [策略实现](https://ml4trading.io/primer/): 27 个主题，从 deflated Sharpe ratio 与层次风险平价（hierarchical risk parity），到 Almgren-Chriss 最优执行。
+- [高级 AI](https://ml4trading.io/primer/): 8 个主题，如马尔可夫决策过程（Markov decision processes）、策略梯度定理（policy-gradient theorem），以及事件预测的正确评分规则（proper scoring rules）。
+- [生产化](https://ml4trading.io/primer/): 2 个主题：冠军-挑战者评估（champion-challenger evaluation），以及特征存储带来的训练-服务偏移（training-serving skew）。
+- [跨章节概念](https://ml4trading.io/primer/): 20 个在各章中引用的构建模块，例如动量与均值回归、偏差-方差权衡（bias-variance tradeoff），以及滚动前向验证（walk-forward validation）。
 
 ### 56 Agent Skills
 
-Reusable, guard-railed tasks for coding agents, each with built-in defenses against lookahead bias, data leakage, and
-multiple-testing errors. Each category links to its full set; a few skills show the range:
+面向编码智能体的可复用、带护栏的任务，每项均内置针对前瞻偏差（lookahead bias）、数据泄露（data leakage）和多重检验错误的防护。每个类别链接到其完整技能集；以下列举部分技能以展示范围：
 
-- [Concepts](https://ml4trading.io/skills/): 10 skills, including lookahead bias, data leakage, and the information
-  coefficient.
-- [Data Acquisition](https://ml4trading.io/skills/): 7 skills spanning fetching data, building bars, and data
-  validation.
-- [Feature Engineering](https://ml4trading.io/skills/): 10 skills, among them computing features, triple-barrier labels,
-  and feature selection.
-- [Evaluation & Validation](https://ml4trading.io/skills/): 8 skills, from walk-forward CV and purging-and-embargo to
-  the deflated Sharpe ratio.
-- [Backtesting](https://ml4trading.io/skills/): 5 skills such as running backtests, cost models, and tear sheets.
-- [Portfolio Management](https://ml4trading.io/skills/): 5 skills, including position sizing, risk metrics, and kill
-  switches.
-- [Infrastructure](https://ml4trading.io/skills/): 4 skills, for example the canonical schema, the registry system, and
-  Polars patterns.
-- [Workflows](https://ml4trading.io/skills/): 5 skills covering factor research, model validation, and production
-  readiness.
-- [Production](https://ml4trading.io/skills/): 2 skills, live trading and monitoring & alerting.
+- [Concepts](https://ml4trading.io/skills/): 10 项技能，涵盖前瞻偏差、数据泄露和信息系数（information coefficient）。
+- [Data Acquisition](https://ml4trading.io/skills/): 7 项技能，涵盖数据获取、构建 bars 与数据验证。
+- [Feature Engineering](https://ml4trading.io/skills/): 10 项技能，包括特征计算、三重障碍标签（triple-barrier labels）和特征选择。
+- [Evaluation & Validation](https://ml4trading.io/skills/): 8 项技能，从步进式交叉验证（walk-forward CV）与清除-禁运（purging-and-embargo）到收缩夏普比率（Deflated Sharpe ratio）。
+- [Backtesting](https://ml4trading.io/skills/): 5 项技能，例如运行回测、成本模型和 tear sheets。
+- [Portfolio Management](https://ml4trading.io/skills/): 5 项技能，包括仓位管理、风险指标和熔断开关（kill switches）。
+- [Infrastructure](https://ml4trading.io/skills/): 4 项技能，例如规范 schema、注册表系统和 Polars 模式。
+- [Workflows](https://ml4trading.io/skills/): 5 项技能，涵盖因子研究、模型验证和生产就绪。
+- [Production](https://ml4trading.io/skills/): 2 项技能：实盘交易与监控与告警。
 
 ---
 
 ## The ML4T Libraries
 
-The notebooks are built on six production Python packages, each documented and usable on its own — one per stage of
-the workflow:
+这些 notebook 基于六个生产级 Python 包构建，每个包均有独立文档且可单独使用——对应工作流的每个阶段：
 
 | Library                                                     | Stage      | What it does                                                                   |
 |-------------------------------------------------------------|------------|--------------------------------------------------------------------------------|
-| [`ml4t-data`](https://ml4trading.io/docs/data/)             | Data       | Unified market-data acquisition from 19+ providers behind one interface        |
-| [`ml4t-engineer`](https://ml4trading.io/docs/engineer/)     | Signal     | Features, labels, alternative bars, and leakage-safe dataset preparation       |
-| [`ml4t-models`](https://ml4trading.io/docs/models/)         | Models     | Finance-native latent factors, SDFs, direct prediction, and portfolio learning |
-| [`ml4t-diagnostic`](https://ml4trading.io/docs/diagnostic/) | Evaluation | Feature validation, strategy diagnostics, and the Deflated Sharpe Ratio        |
-| [`ml4t-backtest`](https://ml4trading.io/docs/backtest/)     | Strategy   | Event-driven backtesting with realistic execution                              |
-| [`ml4t-live`](https://ml4trading.io/docs/live/)             | Deployment | Production trading with broker integrations                                    |
+| [`ml4t-data`](https://ml4trading.io/docs/data/)             | Data       | 通过统一接口从 19+ 家数据提供商获取市场数据                                    |
+| [`ml4t-engineer`](https://ml4trading.io/docs/engineer/)     | Signal     | 特征、标签、替代 bars，以及防泄露的数据集准备                                  |
+| [`ml4t-models`](https://ml4trading.io/docs/models/)         | Models     | 面向金融的潜因子、SDF、直接预测与组合学习                                      |
+| [`ml4t-diagnostic`](https://ml4trading.io/docs/diagnostic/) | Evaluation | 特征验证、策略诊断与收缩夏普比率（Deflated Sharpe Ratio）                      |
+| [`ml4t-backtest`](https://ml4trading.io/docs/backtest/)     | Strategy   | 具备真实istic 执行的事件驱动回测                                               |
+| [`ml4t-live`](https://ml4trading.io/docs/live/)             | Deployment | 对接券商的生产级交易                                                           |
 
 ---
 
-An introduction and a closing chapter bookend six workflow-aligned parts. Chapter titles link to their guides as
-each part is published; the rest are added part by part over the coming weeks.
+导言与收尾章节为六个与工作流对齐的部分作书脊。各章标题链接到对应指南，随各部分陆续发布；其余章节将在未来数周内逐部分添加。
 
 ## Introduction
 
 ### [1. The Process Is Your Edge](01_process_is_edge/)
 
-Why process discipline beats model sophistication. Introduces the ML4T workflow as a research-to-production system,
-regime detection on factor returns and macro indicators, and the evidence boundary that separates exploration from
-confirmation.
+为何流程纪律胜过模型复杂度。将 ML4T 工作流介绍为从研究到生产的系统，涵盖因子收益与宏观指标上的制度识别（regime detection），以及区分探索与确认的 evidence boundary。
 
 ## Part I — Financial Data (Chapters 2–5)
 
-The markets, instruments, and infrastructure the rest of the book builds on: a taxonomy of sources, raw exchange
-messages turned into feature-ready bars, point-in-time fundamentals, and synthetic histories for robust validation.
+全书其余部分所依托的市场、工具与基础设施：数据源分类体系、将原始交易所消息转化为可用于特征的 bars、时点（point-in-time）基本面，以及用于稳健验证的合成历史数据。
 
 ### [2. The Financial Data Universe](02_financial_data_universe/)
 
-A taxonomy of market, fundamental, and alternative data. Surveys eight asset classes, quantifies survivorship bias,
-benchmarks storage formats (Parquet, DuckDB, kdb+, TimescaleDB), and establishes the data-quality framework used
-throughout the book.
+市场、基本面与另类数据的分类体系。调研八类资产，量化幸存者偏差（survivorship bias），对比存储格式（Parquet、DuckDB、kdb+、TimescaleDB），并建立全书沿用的数据质量框架。
 
 ### [3. Market Microstructure](03_market_microstructure/)
 
-From raw exchange messages to feature-ready bars. Parses NASDAQ ITCH, reconstructs limit order books from multiple
-data sources, validates Lee-Ready trade classification, and compares bar-sampling methods — dollar bars deliver the
-best return normality.
+从原始交易所消息到可用于特征的 bars。解析 NASDAQ ITCH，从多数据源重建限价订单簿，验证 Lee-Ready 成交分类，并比较 bar 采样方法——dollar bars 带来最佳的收益正态性。
 
 ### [4. Fundamental and Alternative Data](04_fundamental_alternative_data/)
 
-Point-in-time pipelines for SEC EDGAR filings, entity resolution across identifier systems, macro and commodity
-fundamentals, and alternative-data evaluation — including on-chain crypto fundamentals and prediction markets
-(Kalshi, Polymarket).
+面向 SEC EDGAR 申报文件的时点（point-in-time）流水线、跨标识符体系的实体解析、宏观与大宗商品基本面，以及另类数据评估——包括链上加密基本面与预测市场（Kalshi、Polymarket）。
 
 ### [5. Synthetic Financial Data](05_synthetic_data/)
 
-Generating alternative market histories for robust validation. Implements TimeGAN, Tail-GAN, Sig-CWGAN,
-Diffusion-TS, and LLM-based tabular generation, evaluated through a fidelity–utility–privacy framework.
+生成替代市场历史以进行稳健验证。实现 TimeGAN、Tail-GAN、Sig-CWGAN、Diffusion-TS 与基于 LLM 的表格生成，并通过保真度–效用–隐私框架进行评估。
 
 ## Part II — Research Design and Feature Engineering (Chapters 6–10)
 
-Define the trading problem, then turn data into model-ready signals: research design, labels, features, and the
-evaluation that determines what any model can learn.
+先定义交易问题，再将数据转化为可供模型使用的信号：研究设计、标签、特征，以及决定任何模型能学到什么的评估。
 
 ### [6. Strategy Research Framework](06_strategy_definition/)
 
-Defining the trading game before building models: universe rules, decision schedule, cost model, evaluation
-protocol, and run logging. Introduces the nine case studies and the walk-forward cross-validation discipline that
-anchors Chapters 7–20.
+在建模之前定义交易博弈：标的池规则、决策日程、成本模型、评估协议与运行日志。介绍九个案例研究，以及贯穿第 7–20 章的步进式交叉验证（walk-forward cross-validation）纪律。
 
 ### [7. Defining the Learning Task](07_defining_the_learning_task/)
 
-Label engineering (forward returns, triple-barrier, trend scanning), univariate feature evaluation (information
-coefficients, quantile analysis, feasibility screens), multiple-testing control (BH-FDR, Deflated Sharpe Ratio),
-and causal plausibility checks.
+标签工程（前瞻收益、三重障碍、trend scanning）、单变量特征评估（信息系数、分位数分析、可行性筛选）、多重检验控制（BH-FDR、Deflated Sharpe Ratio）与因果合理性检验。
 
 ### [8. Financial Feature Engineering](08_financial_features/)
 
-Five feature families from price data (momentum, reversal, volatility, liquidity, microstructure), structural and
-cross-instrument features (yield curve, term structure, relative value), contextual features (macro regime, calendar,
-sentiment), and feature selection with robustness testing.
+来自价格数据的五类特征族（动量、反转、波动率、流动性、微观结构），结构与跨工具特征（收益率曲线、期限结构、相对价值），情境特征（宏观制度、日历、情绪），以及带稳健性检验的特征选择。
 
 ### [9. Model-Based Feature Extraction](09_model_based_features/)
 
-Features from fitted models: stationarity diagnostics, Kalman filters, Fourier and wavelet spectral features, GARCH
-volatility, and HMM regime probabilities — with point-in-time correctness enforced throughout.
+来自拟合模型的特征：平稳性诊断、Kalman 滤波器、傅里叶与小波谱特征、GARCH 波动率与 HMM 制度概率——全程强制时点（point-in-time）正确性。
 
 ### [10. Text Feature Engineering](10_text_feature_engineering/)
 
-From bag-of-words through transformers: TF-IDF, Word2Vec and GloVe embeddings, LSTM sequence models, FinBERT
-sentiment, financial NER fine-tuning, and news-return signal construction.
+从词袋到 Transformer：TF-IDF、Word2Vec 与 GloVe 嵌入、LSTM 序列模型、FinBERT 情绪、金融 NER 微调，以及新闻–收益信号构建。
 
 ## Part III — Model Development (Chapters 11–15)
 
-Five model families applied to the same nine case studies, each building on the linear baseline.
+五类模型族应用于相同的九个案例研究，均以线性基线为起点。
 
 ### 11. The ML Pipeline
 
-Regularized linear models (Ridge, LASSO, Elastic Net) as the baseline every later model must beat. Logistic
-regression for direction, SHAP interpretability, conformal prediction for uncertainty, and a cross-dataset
-comparison across all nine case studies.
+正则化线性模型（Ridge、LASSO、Elastic Net）作为后续所有模型必须超越的基线。用于方向的逻辑回归、SHAP 可解释性、用于不确定性的共形预测（conformal prediction），以及跨九个案例研究的数据集对比。
 
 ### 12. Gradient Boosting and Advanced Tabular Models
 
-XGBoost, LightGBM, and CatBoost with Optuna multi-objective tuning, plus deep-learning tabular alternatives (TabPFN,
-TabM). TreeSHAP explainability and cross-dataset results, where gradient boosting is the strongest tabular model in
-most case studies.
+XGBoost、LightGBM 与 CatBoost，配合 Optuna 多目标调参，以及深度学习表格替代方案（TabPFN、TabM）。TreeSHAP 可解释性与跨数据集结果；在多数案例研究中，梯度提升是最强的表格模型。
 
 ### 13. Deep Learning for Time Series
 
-LSTM, N-BEATS, Transformers (PatchTST, iTransformer, TFT), TSMixer, TCN, and Mamba, set against the LTSF-Linear
-debate. A practitioner selection framework and cross-dataset evidence on when deep learning helps and when simpler
-models suffice.
+LSTM、N-BEATS、Transformer（PatchTST、iTransformer、TFT）、TSMixer、TCN 与 Mamba，置于 LTSF-Linear 争论背景下。实践者选型框架与跨数据集证据，说明深度学习何时有帮助、何时更简单的模型已足够。
 
 ### 14. Latent Factor Models
 
-PCA eigenportfolios, IPCA with time-varying loadings, conditional and supervised autoencoders, adversarial SDF
-estimation, and yield-curve decomposition — with cross-dataset results on when latent factors add predictive value.
+PCA 特征组合、带时变载荷的 IPCA、条件与监督自编码器、对抗式 SDF 估计与收益率曲线分解——附跨数据集结果，说明潜因子何时增加预测价值。
 
 ### 15. Causal Machine Learning
 
-Double Machine Learning for isolating factor treatment effects, Bayesian Structural Time Series for event impact, and
-causal discovery (PCMCI, NOTEARS, VAR-LiNGAM), applied across the nine case studies.
+双重机器学习（Double Machine Learning）用于隔离因子处理效应，贝叶斯结构时间序列（Bayesian Structural Time Series）用于事件影响，以及因果发现（PCMCI、NOTEARS、VAR-LiNGAM），应用于九个案例研究。
 
-## Part IV — Strategy Implementation (Chapters 16–20)
+## 第四部分 — 策略实施（第 16–20 章）
 
-From predictions to deployable strategies — backtesting, portfolio construction, costs, risk, and synthesis.
+从预测到可部署策略 — 回测、组合构建、成本、风险与综合。
 
-### 16. Strategy Simulation
+### 16. 策略模拟
 
-Backtesting as falsification: trading-protocol specification, vectorized vs event-driven engines, an ETF baseline
-strategy, core metric reporting, regime diagnostics, and strategy-level overfitting control (Deflated Sharpe Ratio,
-Rademacher Anti-Serum, White's Reality Check).
+将回测视为证伪：交易协议规范、向量化与事件驱动引擎、ETF 基准策略、核心指标报告、市场状态（regime）诊断，以及策略层面的过拟合控制（Deflated Sharpe Ratio、Rademacher Anti-Serum、White's Reality Check）。
 
-### 17. Portfolio Construction
+### 17. 组合构建
 
-From scores to portfolios: mean-variance optimization and its pitfalls, Hierarchical Risk Parity, the Kelly
-criterion, conformal position sizing, deep portfolio allocation, and a controlled allocator comparison across case
-studies.
+从得分到组合：均值-方差优化及其陷阱、层次风险平价（Hierarchical Risk Parity）、Kelly 准则、保形（conformal）仓位 sizing、深度组合配置，以及跨案例研究的受控配置器对比。
 
-### 18. Transaction Costs
+### 18. 交易成本
 
-Cost taxonomy, spread estimation, market-impact calibration, execution algorithms (VWAP, TWAP, Almgren-Chriss
-optimal execution), transaction-cost analysis, and practical guardrails — with breakeven costs that vary widely by
-asset class.
+成本分类、价差估计、市场冲击校准、执行算法（VWAP、TWAP、Almgren-Chriss 最优执行）、交易成本分析，以及实用防护栏 — 盈亏平衡成本因资产类别差异很大。
 
-### 19. Risk Management
+### 19. 风险管理
 
-VaR/CVaR tail measurement, drawdown and path-risk controls, factor and sector decomposition, stress testing,
-adaptive risk overlays, deep hedging, and kill switches. Overlay effectiveness turns out to be strategy-specific.
+VaR/CVaR 尾部度量、回撤与路径风险控制、因子与行业分解、压力测试、自适应风险叠加层、深度对冲（deep hedging）与熔断机制。叠加层的有效性因策略而异。
 
-### 20. Strategy Synthesis
+### 20. 策略综合
 
-What nine experiments reveal about translating ML predictions into strategies: IC–Sharpe decorrelation, Fundamental
-Law diagnostics, the model-family cascade, cost-survival analysis, holdout failure modes, and a practitioner's
-decision framework.
+九项实验揭示的关于将 ML 预测转化为策略的洞见：IC–Sharpe 去相关、基本定律（Fundamental Law）诊断、模型族级联、成本生存分析、留出集（holdout）失效模式，以及面向实践者的决策框架。
 
-## Part V — Advanced AI (Chapters 21–24)
+## 第五部分 — 高级 AI（第 21–24 章）
 
-Reinforcement learning, large language models, knowledge graphs, and autonomous agents for finance.
+强化学习、大语言模型、知识图谱，以及面向金融的自主智能体。
 
-### 21. Reinforcement Learning for Execution and Hedging
+### 21. 面向执行与对冲的强化学习
 
-MDP formulation for finance, DQN/PPO/SAC algorithms, optimal execution, market making with inventory management, deep
-hedging with PFHedge, inverse RL for strategy recovery, and the sim-to-real gap.
+金融领域的 MDP 建模、DQN/PPO/SAC 算法、最优执行、带库存管理的市场做市、基于 PFHedge 的深度对冲、用于策略恢复的逆强化学习（inverse RL），以及仿真到现实（sim-to-real）鸿沟。
 
-### 22. RAG for Financial Research
+### 22. 面向金融研究的 RAG
 
-Retrieval-augmented generation grounded in SEC filings: ingestion, domain-specific embeddings, hybrid retrieval with
-re-ranking, constraint-based prompting, RAG evaluation and failure diagnostics, and the transition to agentic
-workflows.
+基于 SEC 申报文件的检索增强生成（RAG）：摄取、领域专用嵌入、带重排序的混合检索、基于约束的提示、RAG 评估与失效诊断，以及向智能体式工作流的过渡。
 
-### 23. Knowledge Graphs
+### 23. 知识图谱
 
-When graphs earn their infrastructure cost: KG construction from SEC filings, Graph RAG for multi-hop reasoning,
-graph features for ML (GNN embeddings, centrality, community detection), financial networks, and temporal-leakage
-prevention.
+图谱何时值得其基础设施成本：从 SEC 申报文件构建知识图谱（KG）、用于多跳推理的 Graph RAG、面向 ML 的图特征（GNN 嵌入、中心性、社区检测）、金融网络，以及时间泄漏（temporal-leakage）防范。
 
-### 24. Autonomous Agents
+### 24. 自主智能体
 
-Agent architectures (ReAct, Tree of Thoughts, Reflexion), memory systems, tool contracts, the engineering stack
-(LangGraph, Claude SDK), a stateful equity-research agent, multi-agent forecasting with adversarial debate, and
-production reliability.
+智能体架构（ReAct、Tree of Thoughts、Reflexion）、记忆系统、工具契约、工程栈（LangGraph、Claude SDK）、有状态股票研究智能体、带对抗性辩论的多智能体预测，以及生产可靠性。
 
-## Part VI — Production (Chapters 25–26)
+## 第六部分 — 生产部署（第 25–26 章）
 
-Taking strategies live — trading systems and the operational infrastructure that keeps them running.
+将策略上线 — 交易系统与保障其运行的运营基础设施。
 
-### 25. Live Trading Systems
+### 25. 实盘交易系统
 
-A unified framework bridging research and production: Interactive Brokers and Alpaca integration, managed platforms
-(QuantConnect), order-lifecycle management, pipeline verification, and operational readiness.
+连接研究与生产的统一框架：Interactive Brokers 与 Alpaca 集成、托管平台（QuantConnect）、订单生命周期管理、流水线验证，以及运营就绪。
 
-### 26. MLOps and Governance
+### 26. MLOps 与治理
 
-An ML failure taxonomy (pipeline divergence vs performance decay), drift detection, safe model rollout, circuit
-breakers, feature stores, experiment tracking, and the MLOps infrastructure financial ML systems need.
+ML 失效分类（流水线漂移 vs 性能衰减）、漂移检测、安全模型发布、熔断器、特征存储、实验跟踪，以及金融 ML 系统所需的 MLOps 基础设施。
 
-## Conclusion
+## 结语
 
-### 27. The Systematic Edge
+### 27. 系统化优势
 
-The systematic philosophy, quant career paths, learning resources, research frontiers, and how to build your own
-edge. The closing bookend to Chapter 1: the process is the edge.
+系统化理念、量化职业路径、学习资源、研究前沿，以及如何构建你自己的优势。与第 1 章首尾呼应：流程即优势。
 
 ---
 
-## Quick Start
+## 快速入门
 
-These commands are typed into a terminal on your own computer, not into GitHub. New to the command
-line? Start with **[Before You Begin](docs/installation.md#before-you-begin)**.
+以下命令需在你自己电脑的终端中输入，而非在 GitHub 中输入。不熟悉命令行？请从 **[开始之前](docs/installation.md#before-you-begin)** 读起。
 
-Run everything **from the repository root**. Clone and set up with Docker or a local `uv` environment:
+**从仓库根目录运行一切。** 克隆并使用 Docker 或本地 `uv` 环境完成设置：
 
 ```bash
 git clone https://github.com/stefan-jansen/machine-learning-for-trading.git
@@ -366,70 +268,63 @@ docker compose pull ml4t # Option A — Docker (recommended)
 pip install uv && uv sync # Option B — local with uv
 ```
 
-See the **[installation guide](docs/installation.md)** for platform-specific setup (Linux, Windows WSL2, macOS) and GPU
-instructions. Windows readers: WSL2 must be working *before* Docker Desktop is installed, and the
-reboot is not optional.
+平台相关设置（Linux、Windows WSL2、macOS）及 GPU 说明见 **[安装指南](docs/installation.md)**。Windows 读者：必须先让 WSL2 正常工作，*再*安装 Docker Desktop，且重启不可省略。
 
-**Download data.** Most notebooks need datasets; start with the free ones (no API keys):
+**下载数据。** 大多数 notebook 需要数据集；先从免费数据集开始（无需 API 密钥）：
 
 ```bash
 uv run python data/download_all.py --free-only
 ```
 
-The **[data guide](data/README.md)** documents every dataset, API-key setup, the loaders, and storage tiers (≈70 MB
-free tier up to ≈7 GB full).
+**[数据指南](data/README.md)** 记录了每个数据集、API 密钥配置、加载器及存储层级（约 70 MB 免费层至约 7 GB 完整版）。
 
-**Run notebooks.** Notebooks are paired [Jupytext](https://jupytext.readthedocs.io/) files (`.py` source + generated
-`.ipynb`). Run a quick smoke test, or open Jupyter Lab:
+**运行 notebook。** Notebook 与 [Jupytext](https://jupytext.readthedocs.io/) 文件成对（`.py` 源文件 + 生成的 `.ipynb`）。运行快速冒烟测试，或打开 Jupyter Lab：
 
 ```bash
 uv run python 01_process_is_edge/factor_regimes.py
 docker compose up -d ml4t # then open http://localhost:8888
 ```
 
-See the guide to **[running notebooks](docs/running-notebooks.md)** for Papermill parameters and the experiment
-workflow.
+Papermill 参数与实验工作流见 **[运行 notebook 指南](docs/running-notebooks.md)**。
 
-### Docker images
+### Docker 镜像
 
-Most notebooks run on the default **ml4t** image; a few need a specialized one, and each such notebook says so in its
-preamble. Full details in the **[Docker environments guide](envs/README.md)**.
+大多数 notebook 在默认 **ml4t** 镜像上运行；少数需要专用镜像，各 notebook 会在前言中说明。完整说明见 **[Docker 环境指南](envs/README.md)**。
 
 | Image        | Covers                                                           | When you need it        |
 |--------------|------------------------------------------------------------------|-------------------------|
-| `ml4t`       | All 27 chapters + 9 case studies (CPU)                           | Default for everything  |
-| `ml4t-gpu`   | Same `ml4t` image, run with the NVIDIA runtime (`--profile gpu`) | Deep-learning chapters  |
-| `ml4t-py312` | Python 3.12 for signatory, esig, gensim, pfhedge, tfcausalimpact | ~10 notebooks           |
-| `benchmark`  | Database clients (TimescaleDB, ClickHouse, QuestDB, InfluxDB)    | Ch02 storage benchmarks |
-| `rapids`     | RAPIDS cuML + LightGBM CUDA (build locally)                      | One Ch12 GPU benchmark  |
+| `ml4t`       | 全部 27 章 + 9 个案例研究（CPU）                           | 一切内容的默认选择  |
+| `ml4t-gpu`   | 同一 `ml4t` 镜像，使用 NVIDIA runtime（`--profile gpu`）运行 | 深度学习章节  |
+| `ml4t-py312` | Python 3.12，用于 signatory、esig、gensim、pfhedge、tfcausalimpact | 约 10 个 notebook           |
+| `benchmark`  | 数据库客户端（TimescaleDB、ClickHouse、QuestDB、InfluxDB）    | 第 02 章存储基准测试 |
+| `rapids`     | RAPIDS cuML + LightGBM CUDA（本地构建）                      | 第 12 章一项 GPU 基准测试  |
 
 ---
 
-## Releases
+## 版本发布
 
-New chapters and notebooks are added over the coming weeks. ⭐ Watch or star the repo to follow along, and subscribe to
-the twice-weekly [**Insights** newsletter](https://insights.ml4trading.io/).
+新章节与 notebook 将在未来数周内陆续添加。⭐ 关注或 star 本仓库以跟进进展，并订阅每两周一期的 [**Insights** 通讯](https://insights.ml4trading.io/).
 
-**Looking for the second edition?** It is complete and stable on the `second-edition` branch —
-`git checkout second-edition`, and everything is exactly where the book describes it.
+**在找第二版？** 它已在 `second-edition` 分支上完整且稳定 —
+`git checkout second-edition`，一切内容与书中描述完全一致。
 
 ---
 
-## Contributing and Feedback
+## 贡献与反馈
 
-Found an error, a broken link, or have a suggestion? Early feedback is especially valuable before the book launches.
+发现错误、失效链接或有建议？在本书正式发布前，早期反馈尤为宝贵。
 
-- **Issues**: [open a GitHub issue](https://github.com/stefan-jansen/machine-learning-for-trading/issues)
-- **Website and contact**: [ml4trading.io](https://ml4trading.io)
+- **Issues**：[提交 GitHub issue](https://github.com/stefan-jansen/machine-learning-for-trading/issues)
+- **网站与联系**：[ml4trading.io](https://ml4trading.io)
 
 ---
 
-## License
+## 许可证
 
-Code: [MIT License](LICENSE) · Book content: © 2026 Stefan Jansen. All rights reserved.
+代码：[MIT License](LICENSE) · 书籍内容：© 2026 Stefan Jansen. All rights reserved.
 
 <p align="center">
-  <a href="https://amzn.to/4eigy2F">Get the book</a> •
+  <a href="https://amzn.to/4eigy2F">获取本书</a> •
   <a href="https://ml4trading.io">ml4trading.io</a> •
   <a href="https://github.com/stefan-jansen/machine-learning-for-trading">GitHub</a>
 </p>