quantmodel

Research campaign · Iteration 1 · unassessed

RecessionAlert Bear Market High-Risk Alert — public-proxy campaign

Can the incomplete RecessionAlert CMHI Diffusion-zero cash narrative yield transparent, causally timed exploratory public proxies over the longest checked-in history without claiming native source fidelity?

Expressions
7
Logged trials
Not recorded
Independent events
Not assessed
Evidence
unassessed

What the research found

Mechanism and falsifiers

Not recorded in this iteration.

Not recorded in this iteration.

Confidence and limitations

  • The native CMHI Diffusion series and seven paw-print component history are absent; checked-in synthetic breadth and public macro/market proxies are not equivalent inputs.

  • Long-term high/low, moving-average, volume-breadth, Seasonality, Momentum, and Weekly Economic Index formulas are incomplete or undisclosed.

  • Native publication lag, point-in-time availability, release/vintage timestamps, revisions, and calculation clock are unknown; archival fetch delay is not a release lag.

  • The source provides only risk-off cash advice. BUY/re-entry, target weights, instrument, holding period, costs, fills, and same-day timing are not disclosed; overlay sleeves are inferred.

  • The lag-0 control is lookahead-only; causal conclusions use lag 1 or 2. Event windows and returns are descriptive exploratory metrics, not independent validation or investment advice.

  • The macro-confirmed candidate currently duplicates the count proxy and must not be treated as independent evidence.

Not recorded in this iteration.

Compare expressions

Download evidence

Exploratory results. Check each period, proxy and cost assumption before comparing. — means not recorded.

Expression / familyCAGRSharpeMax drawdownTest periodAssessment
cmhi-three-of-seven-lb12-lag1-t0three of seven 9.4%1.08-24.7% Not recorded
 exploratory
cmhi-three-of-seven-lb6-lag1-t0three of seven 6.9%0.80-23.0% Not recorded
 exploratory
cmhi-three-of-seven-lb21-lag1-tneg05three of seven 9.4%0.97-27.2% Not recorded
 exploratory
cmhi-three-of-seven-lb12-lag2-t0three of seven 8.7%0.97-19.1% Not recorded
 exploratory
cmhi-market-internals-lb12-lag1-t0market internals 9.0%0.95-26.4% Not recorded
 exploratory
cmhi-macro-confirmed-lb12-lag1-t0macro confirmed 9.4%1.08-24.7% Not recorded
 exploratory
cmhi-three-of-seven-lb12-lag0-controlthree of seven 19.8%2.03-16.9% Not recorded
 exploratory

Open an expression to inspect its rules and request confirmation. The request must be submitted by a trusted repository collaborator.

cmhi-three-of-seven-lb12-lag1-t0
claim kind

inferred_best_guess

family

three_of_seven

id

cmhi-three-of-seven-lb12-lag1-t0

label

causal best guess

lag bars

1

lookback months

12

threshold

0.0

Confirm as strategyConfirm as indicator
cmhi-three-of-seven-lb6-lag1-t0
claim kind

inferred_nearby_variant

family

three_of_seven

id

cmhi-three-of-seven-lb6-lag1-t0

label

exploratory nearby variant

lag bars

1

lookback months

6

threshold

0.0

Confirm as strategyConfirm as indicator
cmhi-three-of-seven-lb21-lag1-tneg05
claim kind

inferred_nearby_variant

family

three_of_seven

id

cmhi-three-of-seven-lb21-lag1-tneg05

label

exploratory nearby variant

lag bars

1

lookback months

21

threshold

-0.5

Confirm as strategyConfirm as indicator
cmhi-three-of-seven-lb12-lag2-t0
claim kind

inferred_nearby_variant

family

three_of_seven

id

cmhi-three-of-seven-lb12-lag2-t0

label

exploratory nearby variant

lag bars

2

lookback months

12

threshold

0.0

Confirm as strategyConfirm as indicator
cmhi-market-internals-lb12-lag1-t0
claim kind

inferred_family_variant

family

market_internals

id

cmhi-market-internals-lb12-lag1-t0

label

exploratory nearby variant

lag bars

1

lookback months

12

threshold

0.0

Confirm as strategyConfirm as indicator
cmhi-macro-confirmed-lb12-lag1-t0
claim kind

inferred_family_variant

family

macro_confirmed

id

cmhi-macro-confirmed-lb12-lag1-t0

label

exploratory nearby variant

lag bars

1

lookback months

12

threshold

0.0

Confirm as strategyConfirm as indicator
cmhi-three-of-seven-lb12-lag0-control
claim kind

control_only

family

three_of_seven

id

cmhi-three-of-seven-lb12-lag0-control

label

lookahead control

lag bars

0

lookback months

12

threshold

0.0

Confirm as strategyConfirm as indicator

Interactive lab

Interactive history unavailable

Loading available evidence…

Research record

Source claims, inferred rules, experiments and the evidence behind the assessment.

Source
article date

2015-08-25T17:23:00-0400

fetched at

2026-08-22T23:45:17Z

native span

One dated 2015-08-25 Reflections article; narrative windows cover a 20-trading-day deterioration, 2015-08-18 through 2015-08-24 plunge, and named 2015-06/07 component camps. No machine-readable CMHI history is attached.

public boundary

Public summaries, hashes, metadata, and derived proxy metrics only; no cached HTML/assets, credentials, cookies, private data, or unverified native series.

publication lag

Unknown for native CMHI; the 2026-08-22 fetch interval is archival delay, not indicator release lag.

redacted url
https://recessionalert.com/bear-market-high-risk-alert/
sha256

sha256:9bead77031d36f389d1b2204111f926fce443af211023b6ccda3a4febb59000a

title

Bear Market High-Risk Alert | RecessionAlert

What the source claims

  • The article states that CMHI Stock Market Health Diffusion at zero advises zero stock-market exposure / cash positions.

  • Just 3 of 7 major Bear-market paw-prints bearish puts the Diffusion at zero and advises cash positions.

  • The page narrates below-zero new-long-term-high breadth, below-zero NYSE+S&P 500 shares above long-term moving averages, and below-zero net advancing volume breadth.

  • Seasonality is forecast to switch from +1 to -1 on 2015-09-01, creating a fourth bearish paw-print and likely composite-below-zero condition.

  • The source cautions that rapid declines can create transient CMHI-below-zero visits and recommends watching sustained internals, macro, volatility, and SP-500 behavior.

  • No source-linked daily CMHI Diffusion, seven paw-print series, complete formulas, or point-in-time release/vintage history is captured for this row.

  • No explicit BUY, re-entry, instrument, weights, holding period, costs, fill, or same-day execution rule is disclosed.

  • This campaign tests explicitly labeled SPY, synthetic breadth, ^VIX, INDPRO, and UNRATE public analogues; no result is native CMHI evidence.

Rules actually disclosed

  • The article states that CMHI Stock Market Health Diffusion at zero advises zero stock-market exposure / cash positions.

  • Just 3 of 7 major Bear-market paw-prints bearish puts the Diffusion at zero and advises cash positions.

  • The page narrates below-zero new-long-term-high breadth, below-zero NYSE+S&P 500 shares above long-term moving averages, and below-zero net advancing volume breadth.

  • Seasonality is forecast to switch from +1 to -1 on 2015-09-01, creating a fourth bearish paw-print and likely composite-below-zero condition.

  • The source cautions that rapid declines can create transient CMHI-below-zero visits and recommends watching sustained internals, macro, volatility, and SP-500 behavior.

What had to be inferred

  • This campaign tests explicitly labeled SPY, synthetic breadth, ^VIX, INDPRO, and UNRATE public analogues; no result is native CMHI evidence.

Research questions

  • Can the incomplete RecessionAlert CMHI Diffusion-zero cash narrative yield transparent, causally timed exploratory public proxies over the longest checked-in history without claiming native source fidelity?

Data
article published at

2015-08-25T17:23:00-0400

common span

1993-01-31/2026-07-31

inventory path

research/findings/recessionalert_inventory.md

manifest path

research/artifacts/recessionalert/manifest.json

native fidelity

False

proxy database

data/market.sqlite (read-only)

proxy series

  • SPY

  • __SYNTH_SPXBREADTH200__

  • __SYNTH_SPXBREADTH50__

  • ^VIX

  • INDPRO

  • UNRATE

  • IEF

  • BIL

publication lag

Unknown for native CMHI; proxy experiments use month-end latest-available observations plus one or two available-bar lags.

redistribution

Derived proxy values, charts, summaries, hashes, and specs only; no cached page HTML/assets, credentials, cookies, or private account data.

revision policy

Native CMHI revision/vintage policy is undisclosed; local FRED/Yahoo cache provides observation dates only and no source bytes are redistributed.

snapshot database

data/recessionalert.sqlite (read-only)

source fetched at

2026-08-22T23:45:17Z

source page sha256

sha256:9bead77031d36f389d1b2204111f926fce443af211023b6ccda3a4febb59000a

source page url
https://recessionalert.com/bear-market-high-risk-alert/
Baseline implementation
claim kind

inferred_best_guess

family

three_of_seven

id

cmhi-three-of-seven-lb12-lag1-t0

label

causal best guess

lag bars

1

lookback months

12

metrics
cagr

0.09355393452016214

max drawdown

-0.24659452514828128

observations

288

sharpe

1.0766297419218747

sortino

1.646828832800302

total return

7.489233516798109

volatility

0.08654076442809143

threshold

0.0

What to try interactively
  • name

    variant

    note

    Select among inferred proxy families and the lag-0 control; none is native CMHI.

    type

    enum

    values

    • cmhi-three-of-seven-lb12-lag1-t0

    • cmhi-three-of-seven-lb6-lag1-t0

    • cmhi-three-of-seven-lb21-lag1-tneg05

    • cmhi-three-of-seven-lb12-lag2-t0

    • cmhi-market-internals-lb12-lag1-t0

    • cmhi-macro-confirmed-lb12-lag1-t0

    • cmhi-three-of-seven-lb12-lag0-control

  • name

    lookback_months

    note

    Exploratory neighbor for undisclosed long-term windows.

    type

    integer

    values

    • 6

    • 12

    • 21

  • name

    component_threshold

    note

    Exploratory rolling-z-score boundary; source only documents below-zero signs.

    type

    float

    values

    • 0.0

    • -0.5

  • name

    lag_bars

    note

    Lag 1/2 are causal candidates; lag 0 is research-only sensitivity.

    type

    integer

    values

    • 0

    • 1

    • 2

  • name

    risk_off_sleeve

    note

    Inferred overlay sleeves; source only says cash/zero equity.

    type

    enum

    values

    • IEF

    • BIL

Suggested next research

  • Can an authorized checked-in public artifact provide native CMHI daily values, all seven formulas, and release/vintage timestamps?

  • Can the three narrated breadth examples be rebuilt from a declared point-in-time NYSE+S&P 500 constituent universe with exact high/low, MA, and advancing-volume definitions?

  • What publication clock, persistence, reset, holding period, and BUY/re-entry map should replace the inferred overlay semantics?

  • Should a future revision implement a genuinely distinct macro confirmation gate and rerun the full variant/diagnostic matrix?

  • Can IEF and BIL be compared on equal windows with explicit transaction costs before any operational use?

Trial ledger
  • iteration number

    1

    objective

    Map the SHA-verified CMHI narrative, execute labeled public proxy variants, diagnose timing/regime behavior, and publish derived evidence without native indicator admission.

    status

    completed

Agent assessment

  • Publish as exploratory Other/Research evidence; do not register an official CMHI indicator or imply a tradable recommendation.

Combinations

  • Market-only warnings disagree with the count proxy in 43 months; macro-confirmed is exactly duplicated under current implementation.

Historical events

  • The best guess is risk-off around the 2015 source analog, the 2008 grinding-bear analog, and the 2020 selloff; it fails with a -17.90% 2022 analog return.

Parameter sensitivity

  • One-bar and two-bar lag returns are 9.36% and 8.73% annualized versus 19.84% for lag zero; the strict long lookback is nearly flat recently.

Regime behavior

  • Best-guess risk-off occupancy is 57% pre-GFC, 75% during 2008-09, 42% during 2020-21, and 58% in 2022-26; results vary by regime.

State-space exploration

  • Seven variants cover the best guess, nearby lookbacks/thresholds, extra lag, market-only and macro-family neighbors, and a lag-0 lookahead control.

Timing and cost sensitivity

  • No costs are claimed. Lag zero is lookahead-only; all primary evidence uses at least one available-bar lag.

What did not work

  • No tested path establishes native CMHI parity, a source-faithful BUY/re-entry rule, or independent macro confirmation.

What the signal looks like

  • The interactive lab shows the seven tested inferred/proxy signal and equity paths; the charts are not native CMHI observations.

What worked

  • A causal public proxy can be executed over a long checked-in history and approximately catches the dated 2015 event analog.

Why the failures appear to happen

  • The source omits constituent definitions, release clocks, and execution semantics; proxy substitutions and sleeve spans therefore introduce model and timing risk.

Next questions

Review status

published. Research publication does not imply official admission.

Return to pending research