Automating the Generation of Time-Series Trading Strategies

Summary

This Chinese-language excerpt summarizes a report about automatically generating trading strategies from time-series data. It frames conventional factor research, factor combinations, and rule combinations as labor-intensive manual work, then proposes using greater computing capacity to produce candidate strategies in batches. The stated aim is to let users seek strategies aligned with preferred return and risk requirements, making strategy discovery more efficient.

The excerpt provides a high-level motivation and claimed approach, but it does not expose the generation algorithm, data inputs, validation process, portfolio construction, or empirical results. It therefore cannot establish whether the generated strategies outperform manual research or remain robust out of sample. The only explicit caveat is that models built from historical information and data may fail when markets change sharply. Researchers would need the full report and independent tests before drawing conclusions about effectiveness or practical use.

Key ideas

• The report proposes computationally generating many candidate trading strategies from time-series information.
• It aims to reduce the manual effort involved in studying factors and combining trading rules.
• The proposed process is intended to accommodate users' preferred return and risk requirements.
• The excerpt provides no algorithmic details or performance evidence.
• Models based on historical data may break down during sharp market changes.

Tags

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.