Past return order gave no clear gain for five-session Nasdaq Composite 3% fall forecasts in a 2015–2025 archive test

We tested whether past daily return order improves forecasts of a Nasdaq Composite close at least 3% below entry during the next five sessions. We used 537 periods in 2015–2025 and earlier dated ALFRED inputs. The main systems used past residual blocks or all 120 orders of each block. They had the same final-return distribution. Past order gave Brier loss 0.130899, versus 0.131057 for all orders. The gain was 0.000158, with a 95% paired block interval from -0.000546 to +0.000903. Thus this test gives no clear added gain from order. There were 88 path events. Of these, 33 periods ended back above the fall threshold. Final-return events therefore miss some path events. Sixteen planned periods failed an input-age rule. We preserve all forecasts, source records, code, figures, uncertainty draws, and checks. Direct source-derived labels are retained under provider terms. This is one retrospective index test. It does not establish trading profit.

Nasdaq Composite · path risk · filtered historical simulation · return order · dated data · Brier score · negative result