Baa spread inputs gave no clear added weekly Nasdaq risk forecast gain in a 2016–2025 archive test

We tested whether Baa credit spreads improve next-week Nasdaq price risk forecasts after VIX and recent equity risk are known. Dated ALFRED records give 483 fresh test weeks in 2016–2025. Every fit used earlier data and a fixed delay. Adding the spread level and its recent change reduced mean QLIKE by 0.00538. The 95% paired eight-week block interval was -0.03117 to +0.01718. Thus credit inputs gave no clear added gain. Squared error was 22.2% higher, with a wide secondary interval. The small QLIKE gain changed to a loss when 2020 was removed. All planned added-credit intervals included zero. This is a retrospective test of one price index and a fixed ridge model. It does not establish return prediction or trading profit. The package preserves code, dated source requests, hashes, 573 forecasts, complete fitted states, uncertainty draws, figures, checks and reproduction instructions. Provider data remain external under their source terms.

credit spreads · VIX · risk forecasts · Nasdaq Composite · dated data · negative result · out-of-sample forecasts