quantmodel

← Results index

MODELS: Timing entry in new bull markets | RecessionAlert — exploratory proxy campaign

2026-09-03T00:00:00Z Research completed The SHA-pinned timing-entry page documents a 7-tool bear-bottom re-entry toolkit with no executable formulas, named series, vintages, lags, or portfolio mapping. Thirteen labeled public-proxy variants were executed with causal lag>=1 defaults; publication is exploratory Other/Research only with explicit native NO-GO.
1597 recessionalert research_campaign proxy timing-entry fidelity_mismatch data_limited exploratory_tested_proxy uninvestable

Source report: research/campaigns/recessionalert-timing-entry-1597/synthesis.json

Objective

Independently replicate the distinct RecessionAlert inventory row for MODELS: Timing entry in new bull markets from the checked-in data/recessionalert.sqlite snapshot without live-site requests, credentials, private account data, or canonical crawl-database mutation.

Source and provenance

Page https://recessionalert.com/timing-entry-in-new-bull-markets/ bound to SHA256 sha256:cdfea347c81c3ffcfd3ee3592fb7ad666bbf62de2c41ddd0a2e6bcfe4314fc7b, fetched 2026-08-22T23:48:12Z, published 2022-11-09T17:24:17-0500. Public proxies loaded read-only from data/market.sqlite (UNRATE 1948+, DGS10 1962+, DGS3MO 1981+, DFF 1954+, ^GSPC 1927+, SPY/IEF overlay 2002+, breadth 1992+).

Disclosed rule, native span, and publication lag

Documented: BIGBOT 5+ plus ALGO-1/2 re-entry; RAVI 40% to 20th percentile since 1912 monthly; RAVI forecast 30-35% reflex; detrended monthly CMHI -80/0 (-60 variant); FED pivot; UNRATE +2pp; 28-spread >0/>50bp/>100bp. No native observations are pinned and no chart asset bytes are stored on-row (assets=0). Publication lag is undisclosed; causal default is lag>=1 with lag-zero as control_only.

Approach and variants

Primitives in research/research_campaign/recessionalert_1597.py. Thirteen variants across CMHI sigma (1.0/1.5/2.0), RAVI drawdown (35/40%), UNRATE rise (1.5/2.0pp), spread (0/50bp), breadth count (5), forecast band (30-35%), composite trigger-plus-preconditions, and lags (0/1/2) with SPY/IEF overlays.

Results (exploratory)

Composite 448 triggers 5.3% exposure; CMHI 136-173 (~2%); RAVI-dd 341-499; UNRATE 484-648; spread single-gate 6224-7303 (permissive, needs composite); bigbot 1 trigger; forecast 0 triggers (honest negative retained). Diagnostics show UNRATE fires GFC/COVID but silent 2022/recent, spread fires all regimes. Interesting toolkit decomposition with no native parity claim.

Limitations

Native BIGBOT/RAVI/CMHI/ALGO formulas, 28-spread construction, named series, vintages, chart asset bytes, portfolio mapping, and publication clock remain unknown. See research/findings/1597.md and the spec at research/findings/specs/recessionalert-timing-entry-new-bull-markets-1597.json.

Source artifact: research/results/1597.md