Research campaign · Iteration 1 · unassessed
RecessionAlert Weekly Leading Economic Index: exploratory proxy campaign
Can the incomplete RecessionAlert WLEI concept yield a transparent causal exploratory proxy campaign from the longest defensible offline public history without claiming native source fidelity?
- Expressions
- 5
- Logged trials
- Not recorded
- Independent events
- Not assessed
- Evidence
- unassessed
What the research found
The source-backed result is a documented weekly macro growth-index concept, not a playable native indicator: exact observations, formula, threshold, action mapping, native span, and recurring point-in-time release are absent.
Best-effort exploration used checked-in FRED/Yahoo proxy snapshots and a 1993-01-29 to 2026-06-05 weekly display window; the SPY/IEF overlay comparison begins 2002-08-09 because IEF starts then.
Five inferred/proxy variants were executed. Causal weighted variants trail same-window SPY, diffusion is degenerate at its tested threshold, acceleration is interesting but speculative and drawdown-heavy, and same-close gains are non-causal.
This publication preserves the negative, degenerate, timing-sensitive, and candidate results as exploratory evidence; it does not change the native RecessionAlert registry or claim WLEI fidelity.
Mechanism and falsifiers
Not recorded in this iteration.
Not recorded in this iteration.
Confidence and limitations
Native WLEI observations, component membership, formula, weighting, normalization, score range, and source-backed buy/sell mapping are unavailable.
Monthly FRED proxies are carried forward to Friday display bars; this frequency relaxation is explicit and does not reproduce a native weekly release process.
FRED/Yahoo snapshots lack complete point-in-time vintage and release metadata, so causal timing is conservatively represented by a one-available-week execution lag.
The SPY/IEF overlay window starts in 2002 because IEF is unavailable earlier; signal histories are longer but cannot be interpreted as investable overlays without a risk-off sleeve.
The diffusion threshold trial is degenerate and the acceleration result is speculative; no proxy result is an official RecessionAlert WLEI result or catalog admission.
RecessionAlert workbook assets on a separate page are not source-linked to this row and were not merged.
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 |
|---|---|---|---|---|---|
| wlei_proxy_v1weighted-growth | — | 0.44 | -25.7% | Not recorded |
exploratory |
| wlei_proxy_v2weighted-growth | — | 0.59 | -36.5% | Not recorded |
exploratory |
| wlei_proxy_v3diffusion | — | 0.54 | -23.3% | Not recorded |
exploratory failed |
| wlei_proxy_v4acceleration | — | 0.70 | -51.3% | Not recorded |
exploratory candidate |
| wlei_proxy_same_close_sensitivityweighted-growth | — | 0.88 | -29.9% | Not recorded |
research only |
Open an expression to inspect its rules and request confirmation. The request must be submitted by a trusted repository collaborator.
wlei_proxy_v1
- variant id
wlei_proxy_v1
- family id
weighted-growth
- parameters
- lookback
6
- threshold
0.0
- lag bars
1
- holding period
1
- created by
inferred_best_guess
- reason tested
Shortest helper-grid lookback with primary one-available-week causal lag.
- status
exploratory
- tags
interesting_negative_result
parameter_sensitive
wlei_proxy_v2
- variant id
wlei_proxy_v2
- family id
weighted-growth
- parameters
- lookback
12
- threshold
-0.5
- lag bars
1
- holding period
1
- created by
inferred_nearby_variant
- reason tested
Medium lookback and nearby weak threshold with causal lag.
- status
exploratory
- tags
interesting_negative_result
parameter_sensitive
wlei_proxy_v3
- variant id
wlei_proxy_v3
- family id
diffusion
- parameters
- lookback
21
- threshold
0.5
- lag bars
2
- holding period
1
- created by
inferred_nearby_variant
- reason tested
Longer breadth lookback and conservative extra lag.
- status
exploratory_failed
- tags
interesting_negative_result
parameter_sensitive
wlei_proxy_v4
- variant id
wlei_proxy_v4
- family id
acceleration
- parameters
- lookback
12
- threshold
-0.5
- lag bars
1
- holding period
1
- created by
inferred_speculative_extension
- reason tested
Second-difference sensitivity implied by growth emphasis; causal lag retained.
- status
exploratory_candidate
- tags
structurally_interesting
candidate_for_confirmation
needs_more_research
wlei_proxy_same_close_sensitivity
- variant id
wlei_proxy_same_close_sensitivity
- family id
weighted-growth
- parameters
- lookback
6
- threshold
0.0
- lag bars
0
- holding period
1
- created by
inferred_timing_sensitivity
- reason tested
Explicit same-close sensitivity to expose lookahead impact; never used as the causal baseline.
- status
research_only
- tags
interesting_negative_result
fidelity_mismatch
Interactive lab
1741 observationsLoading available evidence…
Research record
Source claims, inferred rules, experiments and the evidence behind the assessment.
Source
- kind
RecessionAlert inventory row
- title
A U.S Weekly Leading Economic Index | RecessionAlert
- sha256
sha256:e81b7ab63069f9cbc94b4a57a526b55b69c32580e11801b1b2a09ba8f806f9fe
- fetched at
2026-08-22T23:45:27Z
- published at
2016-02-23T16:01:55-0500
- content length
86958
- native span
One dated narrative HTML article; no source-linked WLEI observation span
- publication lag
Article claims at most one week of input lag and Thursday-afternoon publication; exact point-in-time release is unknown
- inventory path
research/findings/recessionalert_inventory.md
- manifest path
research/artifacts/recessionalert/manifest.json
- database path
data/recessionalert.sqlite
What the source claims
The cached article describes a U.S. Weekly Leading Economic Index (WLEI) built from more than 20 weekly time series across Corporate Bond, Treasury Bond, Stock Market, Labor Market, Credit Market, and mortgage-backed securities categories.
The article calls WLEI a weekly growth index, says data have at most a one-week lag, and says it is published on Thursday afternoons about 18 hours before the ECRI Weekly Leading Index.
The article gives no source-linked observations, exact component membership, transformation, weights, normalization, lookback, threshold, persistence, state mapping, buy/sell rule, target, sizing, or release vintage policy.
For best-effort exploration, checked-in FRED INDPRO, UNRATE, GS10, and BAA10Y are directionally signed and aligned to Friday bars as a transparent broad proxy; no input is claimed to be a native WLEI component.
Weighted growth, component diffusion, second-difference acceleration, and SPY/IEF risk overlays with explicit lag/lookback/threshold choices are exploratory extensions, not disclosed RecessionAlert rules.
Rules actually disclosed
The cached article describes a U.S. Weekly Leading Economic Index (WLEI) built from more than 20 weekly time series across Corporate Bond, Treasury Bond, Stock Market, Labor Market, Credit Market, and mortgage-backed securities categories.
The article calls WLEI a weekly growth index, says data have at most a one-week lag, and says it is published on Thursday afternoons about 18 hours before the ECRI Weekly Leading Index.
The article gives no source-linked observations, exact component membership, transformation, weights, normalization, lookback, threshold, persistence, state mapping, buy/sell rule, target, sizing, or release vintage policy.
What had to be inferred
For best-effort exploration, checked-in FRED INDPRO, UNRATE, GS10, and BAA10Y are directionally signed and aligned to Friday bars as a transparent broad proxy; no input is claimed to be a native WLEI component.
Weighted growth, component diffusion, second-difference acceleration, and SPY/IEF risk overlays with explicit lag/lookback/threshold choices are exploratory extensions, not disclosed RecessionAlert rules.
Research questions
Can the incomplete RecessionAlert WLEI concept yield a transparent causal exploratory proxy campaign from the longest defensible offline public history without claiming native source fidelity?
Data
- canonical snapshot
data/recessionalert.sqlite pages row for https://recessionalert.com/a-u-s-weekly-leading-economic-index/
- canonical sha256
sha256:e81b7ab63069f9cbc94b4a57a526b55b69c32580e11801b1b2a09ba8f806f9fe
- proxy data
data/market.sqlite read-only FRED INDPRO/UNRATE/GS10/BAA10Y and Yahoo SPY/IEF snapshots
- native spans
WLEI native numeric series unavailable; proxy inputs span 1919-01/2026-08 and broad display window spans 1993-01-29/2026-06-05
- publication lag
Source claims at most one week and Thursday afternoon; primary proxy execution uses one available weekly bar; same-close is research-only.
- revision policy
Checked-in snapshots are fixed for this campaign; macro vintage/revision history is not preserved and no live refresh is performed.
- redistribution
Publish derived proxy values, metadata, metrics, and summaries only; no cached RecessionAlert HTML, chart/image bytes, workbook bytes, credentials, cookies, or private account data.
Baseline implementation
- variant id
wlei_proxy_v1
- family id
weighted-growth
- parameters
- lookback
6
- threshold
0.0
- lag bars
1
- holding period
1
- created by
inferred_best_guess
- reason tested
Shortest helper-grid lookback with primary one-available-week causal lag.
- status
exploratory
- metrics
- cumulative return
2.312456
- annualized return
0.05135
- sharpe
0.440654
- max drawdown
-0.256909
- risk off rate
0.43074
- tags
interesting_negative_result
parameter_sensitive
What to try interactively
- name
threshold
- type
float
- default
0.0
- options
-0.5
0.0
0.5
- name
holding_period
- type
integer
- default
1
- options
1
3
- name
exit
- type
enum
- default
fixed_holding_period
- options
fixed_holding_period
price < sma50
price < sma200
Suggested next research
Can an authorized refresh provide source-linked WLEI observations, exact component membership, and vintage history for this exact row?
Can the Thursday release clock and one-week input lag be confirmed from recurring point-in-time public releases?
Does the v4 acceleration hypothesis survive an untouched-period test with independently versioned data and realistic costs?
What source-backed threshold, state machine, and portfolio action semantics were actually intended?
Can a constituent-level public reconstruction span all six named categories without claiming native parity?
Trial ledger
- iteration number
1
- objective
offline best-effort proxy exploration, diagnostics, and public synthesis
Next questions
Can an authorized refresh provide source-linked WLEI observations, exact component membership, and vintage history for this exact row?
Can the Thursday release clock and one-week input lag be confirmed from recurring point-in-time public releases?
Does the v4 acceleration hypothesis survive an untouched-period test with independently versioned data and realistic costs?
What source-backed threshold, state machine, and portfolio action semantics were actually intended?
Can a constituent-level public reconstruction span all six named categories without claiming native parity?
Review status
published. Research publication does not imply official admission.
Return to pending research