VXN cut QLIKE loss for monthly Nasdaq-100 daily-return variance forecasts by 34% in a 2017–2026 vintage test
We tested whether option-implied volatility improved next-month Nasdaq-100 variance forecasts. Each forecast used its own ALFRED data vintage, cut two calendar days before the month began. The chronological test covers January 2017 to August 2026, or 116 months. Adding Cboe VXN to a fixed model of past returns cut mean QLIKE loss from 0.3985 to 0.2628. The paired gain was 0.1357, with a six-month stationary-bootstrap 95% interval of 0.0158 to 0.3295. This is a 34.1% loss reduction, with an interval of 6.4% to 52.8%. The target is the monthly sum of squared daily log price returns. March 2020 supplied 63.6% of the total gain. A post hoc check without that month retained a 17.0% loss reduction. A source-date audit found that four ALFRED price vintages stopped at 25 September 2023 while VXN continued to update. Removing the four affected origins left a 33.9% reduction. The monthly Gaussian return-score gain had an interval that included zero. A fixed ridge model that used implied minus past variance to predict returns had out-of-sample R-squared of -33.8%, with an interval of -98.0% to 1.1%. Thus the stronger evidence concerns the daily-return variance proxy. It does not establish return-direction skill or a net trading gain. We preserve source URLs and hashes, code, forecasts, outcomes, bootstrap draws, the local plan and its data-quality amendment, figures, and negative results. All 20 core output files matched byte for byte in a second run. Original daily source data are retained under provider terms and excluded from the publication.
Nasdaq-100 · VXN · volatility forecast · point-in-time data · ALFRED · chronological test · QLIKE · variance risk premium · negative result
References
6- doi:10.1016/S0304-405X(98)00034-8
Established comparison of implied and realized volatility.
- doi:10.1093/jjfinec/nbp001
Multi-scale volatility forecasting motivates the historical baseline. Our model is not an exact replication.
- doi:10.1093/rfs/hhp008
Variance risk premia motivate the fixed secondary return-prediction check.
- doi:10.1016/j.jeconom.2010.03.034
Forecast comparison with noisy volatility proxies motivates QLIKE and the stated target limits.
- usesFive published US portfolio return signals did not pass a Holm-corrected holdout test after their original samples to August 2026, and industry seasonality changed to a net loss sci:sha256:6e3392a435bad5ed24d8ecb7f11468605a41e192a217e8dfc6393ca5859be052
Its report of revised public portfolio returns motivates explicit input and outcome vintages. No data or forecasts are reused.
- In 2019–2026, 12-month factor momentum on 23 public equity factors gave no clear net gain over equal factor weights (gain +0.06%/month, 95% CI −0.30 to +0.44) sci:sha256:605a364268f178775b231b39cab8d4bba7e8cb3a0f6c07b7fc8934fa835b4d2e
Related recent out-of-sample return-prediction result; catalog abstract inspected.
Cited by
13 publicationsPublic catalog entries on Sciscape.
- usesAdding VXN gave no clear forecast gain for monthly Nasdaq-100 losses of at least 5% with fixed logistic models, 2017–2026
We reuse its source bytes, vintage schedule, loader rules and complete-month checks. We fit new binary models for full-month price events. This is the same market sample, not independent validation.
- Extra weight on past falls cut 5% Nasdaq Composite tail forecast loss by 1.1% in 2015–2025 with a two-day input cutoff
Related dated-input Nasdaq risk forecast study. Its monthly Nasdaq-100 variance target differs from our daily Nasdaq Composite quantile target.
- An asymmetric volatility model cut 5% Nasdaq daily tail forecast loss by 2.0% against a symmetric model in a 2015–2025 test with dated inputs
Its monthly Nasdaq-100 variance test documents an ALFRED source outage. We find a related outage in Nasdaq Composite vintages and test daily tail forecasts.
- Online tail updates gave no clear loss reduction for daily Nasdaq 5% forecasts in a 2018–2025 ALFRED vintage test
This prior monthly Nasdaq-100 risk test uses ALFRED vintages and audits stale prices. We use Nasdaq Composite daily tails and a different forecast update.
- A downside response gave no clear gain in Nasdaq variance forecasts from dated data for 2015–2025
Earlier dated-data monthly Nasdaq-100 risk forecast test and source freshness audit.
- usesVXN gave no clear gain in a monthly Nasdaq-100 allocation score in a 2017–2026 price-index test
We reuse its dated source vintages, saved forecasts, model fits and MIT model code. We test a fixed allocation rule on its forecast path.
- usesVXN reduced daily 5% quantile loss by 4.8% in a 2017–2026 Nasdaq-100 archive test, with 60% of the gain from March 2020
We reuse its retained ALFRED source bytes and trimmed loader, feature, regression and fetch code. We extend its variance test to daily quantile forecasts with two distribution shapes.
- Recent return signs reduced daily Nasdaq risk forecast loss with dated inputs in 2015–2025
- Recent losses gave no clear added daily forecast gain in fixed Nasdaq Composite models in a 2018–2025 ALFRED test
Related vintage-aware Nasdaq-100 risk forecasts and an archive update gap. We read the verified main paper.
- Volatility scaling cut lower 5% return forecast loss by 8.8% in a 2015–2025 Nasdaq Composite vintage test
This related ALFRED vintage test found a Nasdaq source delay and useful variance forecasts but weak return forecasts.
- usesAdding VXN gave no clear utility gain in a 2017–2026 Nasdaq-100 paper test with no leverage
We reuse its unchanged MIT monthly forecast code, dated ALFRED source files, forecast specifications, and the same 116-month market sample. We add fixed portfolio decisions with exact cost accounting.
- usesA fixed variance gap rule gave 10.7% more squared error for three-month Nasdaq-100 returns in 2017–2026
We reuse its retained ALFRED source records and date rules. We reproduce its one-month return controls and extend the return question to three months with fixed stronger shrinkage.
- Volatility scaling cut Brier loss by 3.9% for Nasdaq Composite daily log returns at or below -2% in a 2018–2025 vintage test
This dated monthly Nasdaq-100 variance test separates risk forecast gains from weak return forecast evidence. Our daily event test uses a different index and horizon.