<p>Comparison of related literature with this study.</p>

Qi Fu's dataset on figshare contains results from a study proposing a Transformer-based Efficiently-Fused Optimized Bayesian Network (Trans-EFOBN) for financial asset pricing. The empirical analysis uses minute-level high-frequency data of CSI 300 constituent stocks from 2019 to 2024, sourced from the Wind database. The dataset, last updated in May 2026, includes performance metrics like a mean absolute error of 0.037 and an R² of 0.86.

Use Cases

• Benchmarking new asset pricing models against the reported MAE of 0.037 and R² of 0.86.
• Analyzing the relationship between macro factors and micro market variables using the described hierarchical structural dependencies.
• Simulating trading strategies based on the reported annualized return of 14.2% and Sharpe ratio of 0.95.
• Studying the application of transformer architectures with temporal logic constraints for sequential feature extraction in finance.

Strengths

• Performance metrics are provided, including a mean absolute error (MAE) of 0.037 and an R² of 0.86.
• The data covers a specific time range (2019-2024) for CSI 300 constituent stocks at a minute-level frequency.
• Model results include simulated trading outcomes with an annualized return of 14.2% and a Sharpe ratio of 0.95.

Limitations

• Column-level documentation is absent; field semantics must be inferred after download.
• Row count is unknown, which may limit suitability assessment.
• The dataset is very small at 9.5 KB, indicating limited scope, likely containing summary results rather than raw data.

Provenance

Source

Wind database, author Qi Fu.

Collection Method

Empirical analysis using a proposed Trans-EFOBN model.

Time Range

2019 to 2024

Freshness

Last updated 2026-05-08 17:28:06; freshness should be verified.

Geography

China (CSI 300 index constituent stocks)