yusufizzetmurat/fomc-volume-har

Artefact for the fed-pulse FOMC text analytics project.

Training corpus

HAR Corsi log-volume regression coefficients + weekday / month-end / quarter-end seasonality block + conformal residual quantiles for horizons h=1, 5, 22. Fit on 180 calendar days of daily ^GSPC volume.

Training command

python -m app.data.late_fusion_volume fit-production-artifact --symbol ^GSPC --period 180d

Attribution

  • Daily share-volume series: Yahoo Finance via yfinance (^GSPC reference symbol).
  • Fit recipe: per-horizon HAR Corsi regression on log-volume with a weekday + month-end / quarter-end seasonality block; conformal residual quantiles at the 80% / 90% bands.
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support