Walk-Forward 实操

Lesson Overview

In "Quant Strategy Development" → "信号与优化", this lesson "Walk-Forward 实操" builds knowledge you can apply immediately.

Walk-Forward按时间多次训练后只在之后区间测试。参数选择与评分应严格分开，标签跨界时需剔除重叠或留gap。 写训练/间隔/测试日历→每折仅用过去选参→冻结→记录下一段表现→拼接不重叠测试，禁止训练收益混入。 本课作业：训练1–60、测试66–80，标签跨度5。最晚训练特征时点是否可为60？说明标签终点条件。

Core Concepts

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

• ▸Walk-Forward按时间多次训练后只在之后区间测试。参数选择与评分应严格分开，标签跨界时需剔除重叠或留gap。
• ▸写训练/间隔/测试日历→每折仅用过去选参→冻结→记录下一段表现→拼接不重叠测试，禁止训练收益混入。
• ▸本课作业：训练1–60、测试66–80，标签跨度5。最晚训练特征时点是否可为60？说明标签终点条件。

Practical Application

After studying, complete the practice tasks below 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 "信号与优化"—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 Walk-Forward 实操 on your own.