Research campaign · Iteration 1 · unassessed
MODELS: The SAHM Rule Redux | RecessionAlert — exploratory proxy campaign
Can the RecessionAlert SAHM Rule Redux MODELS article yield causally timed exploratory insight from labeled UNRATE proxies when native subscriber charts, U2 series, and portfolio mapping are absent?
- Expressions
- 6
- Logged trials
- Not recorded
- Independent events
- Not assessed
- Evidence
- unassessed
What the research found
MODELS SAHM Redux article discloses multiple Sahm constructions but no page-local assets or portfolio map.
Executed 6 labeled UNRATE proxy variants with causal lag>=1 on longest defensible spans.
Highest descriptive causal Sharpe among tested variants: recessionalert.sahm_smoothed.lb12.lag1.t0p5 (0.926).
Disposition exploratory_tested_proxy under Other/Research with explicit NO-GO for native fidelity.
Mechanism and falsifiers
Not recorded in this iteration.
Not recorded in this iteration.
Confidence and limitations
No page-local asset SHA256 or subscriber chart bytes.
U2RATE/state/metro families untestable with current datastore.
SPY/IEF overlays are inferred demonstration harnesses only.
BLS UNRATE revisions and rounding affect unsmoothed triggers.
Results exploratory; not investment advice or catalog admission.
Not recorded in this iteration.
Compare expressions
Download evidenceExploratory results. Check each period, proxy and cost assumption before comparing. — means not recorded.
| Expression / family | CAGR | Sharpe | Max drawdown | Test period | Assessment |
|---|---|---|---|---|---|
| recessionalert.sahm_smoothed.lb12.lag1.t0p5sahm smoothed | 11.5% | 0.93 | -23.9% | 2002-08-31 2026-06-30 |
exploratory |
| recessionalert.sahm_unsmoothed.lb12.lag1.t0p6sahm unsmoothed | 10.3% | 0.84 | -30.4% | 2002-08-31 2026-06-30 |
exploratory |
| recessionalert.sahm_unsmoothed.lb12.lag1.t0p5sahm unsmoothed | 10.7% | 0.89 | -30.4% | 2002-08-31 2026-06-30 |
exploratory |
| recessionalert.sahm_unsmoothed.lb7.lag1.t0p5sahm unsmoothed | 10.7% | 0.85 | -23.9% | 2002-08-31 2026-06-30 |
exploratory |
| recessionalert.sahm_unsmoothed.lb7.lag2.t0p4sahm unsmoothed | 11.2% | 0.88 | -23.9% | 2002-08-31 2026-06-30 |
exploratory |
| recessionalert.sahm_smoothed.lb12.lag2.t0p4sahm smoothed | 10.5% | 0.87 | -23.9% | 2002-08-31 2026-06-30 |
exploratory |
Open an expression to inspect its rules and request confirmation. The request must be submitted by a trusted repository collaborator.
recessionalert.sahm_smoothed.lb12.lag1.t0p5
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
sahm_smoothed
- id
recessionalert.sahm_smoothed.lb12.lag1.t0p5
- lag bars
1
- lookback
12
- smoothed
True
- source fidelity
False
- threshold
0.5
recessionalert.sahm_unsmoothed.lb12.lag1.t0p6
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
sahm_unsmoothed
- id
recessionalert.sahm_unsmoothed.lb12.lag1.t0p6
- lag bars
1
- lookback
12
- smoothed
False
- source fidelity
False
- threshold
0.6
recessionalert.sahm_unsmoothed.lb12.lag1.t0p5
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
sahm_unsmoothed
- id
recessionalert.sahm_unsmoothed.lb12.lag1.t0p5
- lag bars
1
- lookback
12
- smoothed
False
- source fidelity
False
- threshold
0.5
recessionalert.sahm_unsmoothed.lb7.lag1.t0p5
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
sahm_unsmoothed
- id
recessionalert.sahm_unsmoothed.lb7.lag1.t0p5
- lag bars
1
- lookback
7
- smoothed
False
- source fidelity
False
- threshold
0.5
recessionalert.sahm_unsmoothed.lb7.lag2.t0p4
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
sahm_unsmoothed
- id
recessionalert.sahm_unsmoothed.lb7.lag2.t0p4
- lag bars
2
- lookback
7
- smoothed
False
- source fidelity
False
- threshold
0.4
recessionalert.sahm_smoothed.lb12.lag2.t0p4
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
sahm_smoothed
- id
recessionalert.sahm_smoothed.lb12.lag2.t0p4
- lag bars
2
- lookback
12
- smoothed
True
- source fidelity
False
- threshold
0.4
Interactive lab
Interactive history unavailableLoading available evidence…
Research record
Source claims, inferred rules, experiments and the evidence behind the assessment.
Source
- fetched at
2026-08-22T23:48:25Z
- native span
single 2024-04-01 HTML snapshot; no source-linked subscriber chart or xlsx asset
- publication lag
monthly labor-report cadence qualitative; per-variant release clock undisclosed
- redacted url
- https://recessionalert.com/sahm-rule-redux/
- sha256
sha256:9e9dc30d8e2d67ec4b12fd7b31d0589fb92659e575f105def7c6290949ab434e
- title
MODELS: The SAHM Rule Redux | RecessionAlert
What the source claims
Standard Sahm: 3MMA(UNRATE) minus prior 12m minimum; trigger 0.50.
Redux #1: unsmoothed vs 12m min; trigger 0.60 to address June 2003 false positive.
Redux #2: unsmoothed vs prior 7m min; eliminates June 2003 false positive.
Cycle-low Schannepp lookback beyond 12m to cycle trough (examples 13–19 months).
State individualized Sahm triggers 0.4%–1.0% per state; U2 Job Losers Redux variant.
Monthly subscriber chart cadence; no page-local xlsx/csv/chart asset on this row.
UNRATE public proxies for undisclosed native microdata; SPY/IEF/BIL overlays for demonstration only.
U2RATE absent from datastore; state/metro enumeration and subscriber chart byte parity unavailable.
Rules actually disclosed
Standard Sahm: 3MMA(UNRATE) minus prior 12m minimum; trigger 0.50.
Redux #1: unsmoothed vs 12m min; trigger 0.60 to address June 2003 false positive.
Redux #2: unsmoothed vs prior 7m min; eliminates June 2003 false positive.
Cycle-low Schannepp lookback beyond 12m to cycle trough (examples 13–19 months).
State individualized Sahm triggers 0.4%–1.0% per state; U2 Job Losers Redux variant.
Monthly subscriber chart cadence; no page-local xlsx/csv/chart asset on this row.
What had to be inferred
UNRATE public proxies for undisclosed native microdata; SPY/IEF/BIL overlays for demonstration only.
Research questions
Can the RecessionAlert SAHM Rule Redux MODELS article yield causally timed exploratory insight from labeled UNRATE proxies when native subscriber charts, U2 series, and portfolio mapping are absent?
Data
- database
data/market.sqlite read-only
- fetched at
2026-08-22T23:48:25Z
- frequency
month_end
- page sha256
sha256:9e9dc30d8e2d67ec4b12fd7b31d0589fb92659e575f105def7c6290949ab434e
- proxies
UNRATE
SPY
IEF
BIL
- ra database
data/recessionalert.sqlite read-only; no mutation
Baseline implementation
- claim kind
inferred
- claim role
causal_primary
- exploratory
True
- family
sahm_smoothed
- id
recessionalert.sahm_smoothed.lb12.lag1.t0p5
- lag bars
1
- lookback
12
- metrics
- cagr
0.11459193766885756
- end
2026-06-30
- max drawdown
-0.23927178828328344
- n months
286
- risk off events
15
- risk on share
0.771551724137931
- sharpe
0.925855479023602
- start
2002-08-31
- vol
0.12626878298736682
- smoothed
True
- source fidelity
False
- threshold
0.5
What to try interactively
- name
family
- type
enum
- values
sahm_smoothed
sahm_unsmoothed
- name
lag_bars
- type
integer
- values
0
1
2
- name
lookback
- type
integer
- values
7
12
- name
threshold
- type
number
- values
0.4
0.5
0.6
Suggested next research
Can an authorized refresh attach subscriber chart SHA256 assets?
Does importing U2RATE enable Job Losers Redux testing?
Would first-release macro vintages change causal lag conclusions?
Trial ledger
- iteration
1
- note
Initial offline best-effort proxy campaign for issue 1565
- status
published
Next questions
Can an authorized refresh attach subscriber chart SHA256 assets?
Does importing U2RATE enable Job Losers Redux testing?
Would first-release macro vintages change causal lag conclusions?
Review status
published. Research publication does not imply official admission.
Return to pending research