Cannot be evaluated on obtainable data. Requires historical option-implied volatility surface for agricultural futures.
Not tested. The data required does not exist in obtainable form, so no result is claimed in either direction.
Skew/variance premium in ag options around ENSO onsets
The hypothesis is not whether realised volatility rises around ENSO onsets — that was tested and returned noise (3 hits in 39 tests). It is whether realised volatility exceeds what options were pricing at the time. That requires a historical implied-vol surface. yfinance exposes only the current option chain with no history, so the variance risk premium cannot be measured at any point in the past.
Publishing this as a verdict rather than dropping it keeps the denominator honest: a hypothesis that was never tested must not be mistaken for one that failed.
| Field | Detail |
|---|---|
| Requires | Historical option-implied volatility surface for agricultural futures |
| Reason | The hypothesis is not whether realised volatility rises around ENSO onsets — that was tested and returned noise (3 hits in 39 tests). It is whether realised volatility exceeds what options were pricing at the time. That requires a historical implied-vol surface. yfinance exposes only the current option chain with no history, so the variance risk premium cannot be measured at any point in the past. |
| Where it exists | Historical option-implied volatility surfaces are distributed by OptionMetrics (IvyDB), CME DataMine and Barchart. Availability and terms not assessed here. |
| Rev | Stage | Status | Change |
|---|---|---|---|
| 01 | Draft | — | Direction and criteria fixed before the experiment ran |
| 02 | Assessed | Disqualified | Required data is not obtainable; recorded rather than dropped |