Skip to content

Research

Every measurement, in full

This engine writes 89 read-only artifacts and publishes all of them. Several are corrections against our own earlier numbers and several are nulls: results that closed a line of enquiry rather than opening one. Those are the entries worth reading first, because a record that only contains its wins is not a record.

Each page below is generated from the artifact itself. No figure on any of them is typed by hand, so a number that moves in the engine moves here, and a number that exists nowhere in an artifact cannot appear at all.

The measurements

  • Active ownership blind handoff receipt

    The archive is ready for a genuinely independent source review. It contains no machine predictions or return data and does not itself provide accuracy evidence.

  • Active ownership confirmatory design

    This receipt freezes a future disjoint classifier-confirmation design.

  • Active ownership human gate audit

    This is a prospective gate-design and reachability audit. It uses frozen machine outputs but no human outcomes, prices, returns, or portfolio results.

  • Admission dry run / Result

    Admissibility MECHANICS on already-registered identities. Registers no hypothesis identity, proposes no sleeve, and makes no claim about any candidate's edge.

  • Admission v7 promotion / Receipt

    Admission v7 promotion / Receipt: a read-only artifact regenerated from the engine run that produced it. The figure and its limits are published together.

  • Alpaca broker reconciliation

    Self-published read-only reconciliation against three dedicated Alpaca paper accounts.

  • Alphavintage sealed outcome

    The figures the published correction paper quotes, recomputed from the probe's own artifacts so a reader can check them.

  • Atlas reachability screen

    Screens obtainability only. Opens no return data, registers no hypothesis identity, runs no backtest, and claims nothing about any family's edge, sign,…

  • Author protocol review packets

    These are blank hash-bound protocol review packets. They prove no author answers, approval, identity, next-stage authorization, external review, or publication.

  • Bond etf nav reachability

    This classifies ownership of four failed source gates. It does not claim that paid data will pass them, opens no market records or returns, and makes no…

  • Book without alphavintage

    A measurement of two allocations on one common window, both reported. Opens no new data, registers no hypothesis identity, changes no allocation and…

  • Cftc release reachability

    This is a reachability audit over metadata only. It assumes the best possible exact coverage for the 13 months CFTC says are available, opens no position…

  • Claim coverage map

    Maps each PUBLISHED ARTIFACT to the mechanisms that guard it and records when each mechanism last ran, by running it.

  • Contract and unit audit

    Three audit dimensions worked against the contract in force and the published bundle.

  • Cost model realism

    Reads the live trading databases read-only. Registers no hypothesis identity, opens no return data, and changes no cost parameter. 0 trials.

  • Crypto lab carry crash incident

    Forensic description of one paper-traded LABUSDT episode. It spends zero return hypotheses, does not estimate Sharpe or expected performance, does not…

  • Crypto position attribution

    Latest-cycle arithmetic attribution for the locally simulated crypto paper account.

  • Crypto position attribution preflight observation

    This receipt proves only a successful read-only preflight at the stated instant.

  • Crypto position attribution rollout verification

    This verifies only prospective position-attribution operation on the first eligible natural cycle.

  • Current book diversification

    Retrospective diversification measurement of the exact current three-sleeve research specification and strategic overlay.

  • Current book drawdown

    This is a zero-drift two-year risk simulation of the exact current three-sleeve fixed-weight research composition and strategic overlay.

  • Data gate unblocks

    Classifies blockers only. Opens no return data, runs no backtest, registers no hypothesis, and claims nothing about any candidate's edge, sign or correlation.

  • Data lake scale

    Counts what is on disk and what pytest collects. It says nothing about data quality, coverage gaps, or whether any of it supports a trading claim; only how…

  • Drawdown live estimator

    A simulation through a vectorized twin of the PRODUCTION covariance estimator, asserted equal to ewma_cov on real states before any result was reported.

  • Execution gap power

    Reads trading databases and walk-forward curves read-only. Registers no hypothesis identity and opens no return data. 0 trials.

  • External publication

    These are incomplete preparation bundles. No external submission, DOI, peer review, citation, moderation outcome, or independent replication is claimed.

  • External validation opportunities

    This is a dated official-source opportunity audit. It proves no personal eligibility, registration, submission, judging, advancement, award, review,…

  • Feasibility gate reachability

    Asks only whether each near-miss gate is reachable by repairing extraction.

  • Forward drawdown evidence

    The sealed study cell is a 96-path, two-year, fourteen-sleeve, daily simulation at a 10% book-level volatility target.

  • Forward evidence contract

    This contract freezes how forward evidence will be described before the record is mature.

  • Forward evidence maturity

    Paper-only, self-published evidence. An immature or observed Sharpe is not an established Sharpe.

  • Forward sleeve contribution

    Descriptive arithmetic attribution of the published paper curve only. It is not a Sharpe estimate, evidence of expected sleeve performance, permission to…

  • Guards that cannot fire

    A STRUCTURAL scan of source for checks that cannot fail. It reads code, not results: it cannot say whether a guard is correct, only whether it is capable of…

  • Hdb dividend vendor resolution

    This GET-only vendor audit identifies the Sharadar June-26 zero row as a due-bill administrative date associated with a separate positive August-11 HDB ADS…

  • Identity redesign / Spinoff form universe

    Counts filings by form type and enumerates the action types held in the corporate-action lake.

  • Ledoit wolf effective sample

    Measures an approximation's error on real price data. Changes nothing, registers no hypothesis identity, and opens no return data on any untested identity.

  • Live covariance memory

    A property of the ESTIMATOR, derived from its exact weighting and verified against the production function by unit impulse.

  • Merger announcement confirmatory design

    Freezes a disjoint no-return confirmation design for a materially new announcement identity.

  • Mutation ledger

    Breaks what each guard watches and records whether the guard failed. Runs no backtest, opens no return data, registers no hypothesis identity, and changes…

  • Next sleeve selection

    This selects the next feasibility candidate using source reachability and mechanism criteria only.

  • Operating margin corrected replay authorization

    This authorizes one zero-trial corrected-data reproduction of the immutable single_operating_margin identity. The global lake split gate remains failed.

  • Operating margin corrected reproduction

    The corporate-action correction materially changes the measured curve but does not rescue the factor: corrected Sharpe, CAGR, and total return remain negative.

  • Operating margin exposed split resolution

    This authorizes bypassing only the price-gap heuristic for the named event, date, and ratio tuples.

  • Operating margin replay infrastructure failure

    This records a physical data-layout failure before signal or return computation.

  • Operating margin split exposure

    This reads only sealed order and position artifacts, not equity or return series.

  • Orthogonality prior

    A PRIOR, stated before measurement and labelled as such in every row. No candidate correlation is measured, no hypothesis identity is registered, no return…

  • Polygon split crosscheck

    This GET-only cross-check measures external corporate-action agreement.

  • Portfolio evidence

    This is factual portfolio evidence, not a Stanford admissions claim or endorsement.

  • Prereg earnings narrative parameters

    The numeric parameters this pre-registration commits to, extracted from the committed document and stated as data.

  • Program status

    This contract states the programme's governing targets, current paper record, execution provenance and evidence coverage.

  • Prospective trial record

    This receipt seals one preregistered historical walk-forward simulation and its same-path diagnostics.

  • Reachability harness

    Reads frozen feasibility artifacts read-only. Registers no hypothesis identity, opens no return data, and authorises no candidate. 0 trials.

  • Record continuity

    Reads the per-sleeve trading databases read-only. Opens no market data, registers no hypothesis identity and changes nothing.

  • Repurchase issuance companyfacts audit

    Direct-tag coverage cannot pass the full family feasibility gate. Item 703 table precision/recall and contamination reconciliation remain mandatory.

  • Repurchase issuance identity overlap audit

    Completed flows may be a cleaner measurement, but cleanliness is not economic independence.

  • Repurchase issuance semantics audit

    Repurchase issuance semantics audit: a read-only artifact regenerated from the engine run that produced it. The figure and its limits are published together.

  • Research accessibility / Audit

    This receipt seals Lighthouse and supplemental automated Chromium accessibility evidence across the four named public routes on local production builds.

  • Sharadar corporate action corrected lake

    This proves only that a new physical data version was materialized under the sealed source transformation contract.

  • Sharadar corrected corporate action validation

    The dividend normalization passes globally, but the lake fails the independent split gate.

  • Sharadar dividend basis resolution

    This authorizes a new physical data version for validation only. The adjustment-basis interpretation is an issuer-anchored inference because public ACTIONS…

  • Sharadar dividend price consistency

    This is a source-data and executable-semantics audit. It opens no signals, positions, returns, Sharpe ratios, or strategy outcomes and makes no investment…

  • Sharadar dividend split basis

    The arithmetic strongly supports a split-adjustment-basis explanation for 421 rows, but public vendor documentation labels ACTIONS.value only as numeric.

  • Sharadar hdb corrected lake

    This artifact proves a new physical Sharadar lake version differs only by explicit quarantine of the exact HDB zero-cash due-bill marker.

  • Sharadar split governance policy

    This closes the governance-routing gap, not the global data-quality gate.

  • Sharadar split lifecycle scope

    Lifecycle classification does not delete source events or authorize ratio repair.

  • Sharadar zero dividend quarantine

    This audit proves that one of 11 vendor-originated non-positive cash-dividend rows propagated unchanged into the filtered Sharadar executable lake and that…

  • Sleeve admission contract / Contract

    This in-force contract applies only to return identities reserved at ordinal 229 or later under its exact effective hash.

  • Sleeve atlas / Atlas

    Sleeve atlas / Atlas: a read-only artifact regenerated from the engine run that produced it. The figure and its limits are published together. The raw JSON is…

  • Sleeve atlas / Audit

    Passing this audit is not return evidence and does not create a sleeve. It only permits a separately sealed key-free feasibility review.

  • Sleeve atlas / Sleeve family lineage audit

    This audit proves registry consistency against the enumerated internal ledgers. It does not prove global economic novelty, returns, or sleeve admission.

  • Sleeve discovery

    Sleeve discovery: a read-only artifact regenerated from the engine run that produced it. The figure and its limits are published together. The raw JSON is one…

  • Sleeve quality decomposition

    A FIRST-ORDER decomposition from published artifacts, read-only. Each dollar component is converted to an annualised return on starting equity and then to…

  • Spinoff prorata gate

    A measurement of document LANGUAGE on 98 frozen Form 10 filings, every one hash-verified against the feasibility artifact.

  • Split exception issuer resolution

    Issuer evidence resolves only these named exceptions. It does not authorize a strategy replay, alter a historical result, or validate performance.

  • Split issuer conflict resolution batch v5

    This identifies eleven issuer-date, duplicate, or composite mismatches, four ratio-semantics conflicts, and one issuer-identity/timeline conflict.

  • Split issuer resolution batch v10

    This seal verifies two McLeodUSA splits and the issuer-effective NCI1 split.

  • Split issuer resolution batch v11

    This seal verifies the exact BYNDQ, NWKC, XCEDQ, CB, and CCIL share mutations. It does not authorize another event, open returns, or pass the global split gate.

  • Split issuer resolution batch v2

    Issuer evidence resolves only AMPE on its first resumed split-adjusted trading date and EVHC as a merger exchange conversion on its first Newco trading date.

  • Split issuer resolution batch v3

    This seal verifies the old-share 1-for-9,200 mutation and its first frozen post-effective-date bar.

  • Split issuer resolution batch v4

    This verifies only the ATNI, CMO, ITT, SBRA, NRF, IHG, KMI1, KSU, TYC, T1, RBAK, and SDH1 share-mutation components.

  • Split issuer resolution batch v6

    This seal verifies one 1-for-8 reverse split and its first frozen post-effective-date bar.

  • Split issuer resolution batch v7

    This seal verifies six plain splits at their issuer-bound frozen bars: BNI, EPAC, GDW, HAFC, JAKK, and LNG.

  • Split issuer resolution batch v8

    This seal verifies four plain split events for CRGN, ICIX, and OATS. It does not open returns, authorize another event, or pass the global split gate.

  • Split issuer resolution batch v9

    This seal verifies only UTI1's plain two-for-one split. It does not open returns, authorize another event, or pass the global split gate.

  • Split lifecycle discontinuity resolution

    The issuer filings and frozen price sequence establish cancellation/restructuring boundaries involving new equity, aggregate allocations, and/or warrants.

  • Tender offer reachability

    This audit classifies whether parser work is evidentially justified. It does not claim edge, sign, return, Sharpe, drawdown, capacity, correlation or admission.

  • Trial accounting

    This ledger proves what was measured and how selection N is counted. It does not prove that any identity is profitable, independent, or admissible.

  • Unresolved split event context

    Nearby context is descriptive evidence, not causal proof. No source row is deleted, no ratio is changed, and no replay or performance claim is authorized.

  • Vate 2020 dividend resolution

    This authorizes only exact-row quarantine in a new physical lake version.

Where these come from

Each is regenerated by the engine that produced the result and copied verbatim to the glass box, where a downloadable verifier re-checks every content hash and signature against the published bundle. The written-up versions live in the research library.