Figures Data from Fig 8–Fig 14: CSI 300 High-Frequency Trading Model Results

Figures data from Fig 8–Fig 14, supporting a study on a Transformer-based Efficiently-Fused Optimized Bayesian Network (Trans-EFOBN) for financial asset pricing. The dataset, authored by Qi Fu and last updated in May 2026, contains empirical results from minute-level high-frequency data of CSI 300 constituent stocks from 2019 to 2024. The study reports a mean absolute error of 0.037 and an R² of 0.86 for the proposed model.

Use Cases

• Benchmarking new asset pricing models against reported MAE and R² metrics.
• Analyzing the performance of transformer-Bayesian network hybrids in simulated trading scenarios.
• Studying the relationship between macro factors and micro market variables in high-frequency contexts.

Strengths

• Model performance metrics are provided, including a mean absolute error of 0.037 and an R² of 0.86.
• The underlying study uses minute-level high-frequency data from 2019 to 2024, suggesting a detailed temporal scope.
• The dataset is openly licensed under CC-BY-4.0.

Limitations

• Description metadata is limited; actual data quality requires manual inspection after download.
• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.

Provenance

Source

Wind database, as cited in the study description.

Collection Method

Empirical analysis using minute-level high-frequency data.

Time Range

2019 to 2024

Freshness

Last updated 2026-05-08 17:27:48; freshness should be verified.

Geography

China (CSI 300 index constituent stocks)