金融 ML 特征

Lesson Overview

In "Machine Learning for Trading" → "特征工程", this lesson "金融 ML 特征" builds knowledge you can apply immediately.

金融ML特征要在预测时点可获得。收益、动量和滚动波动是价格变换；样本单位、频率及缺失处理须明确，不能含未来标签或全样本标准化结果。 给每列写公式/窗口/可用时点→按时间拆分→只在训练拟合处理→删除起始缺值→核特征标签隔离。 本课作业：解释r1、momentum3、vol5三列分别需要哪些历史行，哪些起始行不能训练。

Core Concepts

These points align with the video walkthrough—take notes as you watch.

• ▸金融ML特征要在预测时点可获得。收益、动量和滚动波动是价格变换；样本单位、频率及缺失处理须明确，不能含未来标签或全样本标准化结果。
• ▸给每列写公式/窗口/可用时点→按时间拆分→只在训练拟合处理→删除起始缺值→核特征标签隔离。
• ▸本课作业：解释r1、momentum3、vol5三列分别需要哪些历史行，哪些起始行不能训练。

Practical Application

After studying, complete the companion practice lesson to move from understanding to doing.

• ▸Follow along in Jupyter or VS Code
• ▸Log how parameter changes affect backtest results
• ▸Submit a reproducible analysis notebook

FAQ

• ▸Q: How does this lesson fit the course? A: Part of "特征工程"—follow the curriculum order.
• ▸Q: What do I need? A: Demo account and notes; quant lessons need a Python environment.
• ▸Q: How do I know I've mastered it? A: Pass the quiz and explain three key points of 金融 ML 特征 on your own.