Factor Discovery on Tushare Pro A-share (2022-2026)
49.8YieldHybrid
Evaluation results
| Method | Links | ||
|---|---|---|---|
| HybridSubset=High-corr, reff/N=0.0032026.06 | 49.8 | 11 | |
| StructuredSubset=High-corr, reff/N=0.0032026.06 | 38.8 | — | |
| HybridSubset=All factors, reff/N=0.0032026.06 | 31.2 | 6 | |
| HybridSubset=Random 50%, reff/N=0.0062026.06 | 29.8 | 6.5 | |
| HybridSubset=Decorrelated, reff/N=0.0862026.06 | 26.4 | 3.1 | |
| StructuredSubset=All factors, reff/N=0.0032026.06 | 25.2 | — | |
| StructuredSubset=Random 50%, reff/N=0.0062026.06 | 23.3 | — | |
| StructuredSubset=Decorrelated, reff/N=0.0862026.06 | 23.2 | — | |
| HybridSubset=Low-corr, reff/N=0.0252026.06 | 22.7 | 0.8 | |
| StructuredSubset=Low-corr, reff/N=0.0252026.06 | 21.9 | — | |
| LLM-styleSubset=High-corr, reff/N=0.0032026.06 | 9.1 | — | |
| LLM-styleSubset=Random 50%, reff/N=0.0062026.06 | 5.2 | — | |
| LLM-styleSubset=All factors, reff/N=0.0032026.06 | 5 | — | |
| LLM-styleSubset=Decorrelated, reff/N=0.0862026.06 | 2.6 | — | |
| LLM-styleSubset=Low-corr, reff/N=0.0252026.06 | 0.6 | — | |
| RandomSubset=All factors, reff/N=0.0032026.06 | 0 | — | |
| RandomSubset=High-corr, reff/N=0.0032026.06 | 0 | — | |
| RandomSubset=Random 50%, reff/N=0.0062026.06 | 0 | — | |
| RandomSubset=Low-corr, reff/N=0.0252026.06 | 0 | — | |
| RandomSubset=Decorrelated, reff/N=0.0862026.06 | 0 | — |