--- license: other pretty_name: vol-forecast data (pinned 2026-09) tags: [finance, volatility, time-series, tabular] --- # vol-forecast data — pinned version `v2026.09` Inputs for forecasting next week's realised volatility of the US stock market. Code: https://github.com/StevenLi-phoenix/vol-forecast | File | Source | |---|---| | `F-F_Research_Data_5_Factors_2x3_daily_CSV.zip`, `F-F_Momentum_Factor_daily_CSV.zip` | [Kenneth R. French Data Library](https://mba.tuck.dartmouth.edu/pages/faculty/ken.french/Data_Library/) (CRSP 202609 release), unmodified | | `VIXCLS.csv`, `T10Y3M.csv`, `BAA10Y.csv` | [FRED](https://fred.stlouisfed.org/), unmodified, downloaded 2026-11-01 | | `weekly.csv` | derived: 1893 weekly rows (1990-03-23 .. 2026-09-25), 14 core + 2 optional features, target `target_lrv_next` | | `MANIFEST.json` | sha256 and size of every file above | The upstream data belong to their providers; check their terms before redistributing. Rebuild `weekly.csv` with `scripts/build_dataset.py` in the code repo; verify files against `MANIFEST.json`.