Equity Risk Premium Forecast Performance Metrics by Method and Threshold

A table reporting in-sample and out-of-sample R² measures for equity risk premium forecasts, categorized by feature representation, estimation method, and predictor threshold. The dataset includes separate performance metrics for downside and upside market months, as defined in the source publication. Authored by Jeonggyu Huh and last updated on May 15, 2026.

Use Cases

• Benchmarking forecasting models based on out-of-sample R² performance relative to a historical mean benchmark.
• Evaluating the impact of different feature representation and estimation methods on prediction accuracy.
• Analyzing forecast performance in specific market conditions based on downside and upside tail month metrics.
• Determining optimal predictor inclusion thresholds for combined feature sets in multi-stage forecasting pipelines.

Strengths

• Dataset is openly licensed under CC-BY-4.0.
• Includes specific performance metrics like in-sample R² and three out-of-sample R² measures.
• Distinguishes between downside and upside market month performance.

Limitations

• Dataset is very small at 9.5 KB, indicating limited scope.
• Row count and column-level documentation are unknown, requiring manual inspection after download.
• Description metadata is limited; actual data quality requires manual inspection after download.

Provenance

Source

figshare

Freshness

Last updated 2026-05-15 17:56:32; freshness should be verified.