Historical market research · evidence firstQuant Market Research Archive

A public archive of historical market research, with dates, progress, and limitations made explicit.

Public research · see dates belowAggregate results · historical description only
Data and versions

Every public snapshot has its own dates and boundaries

Generation time is not the end of data coverage. Missing dates remain unspecified; build time is not used as a substitute.

Selected key public evidence files are linked below. Individual experiments retain their own dates and additional sources in the report. A fingerprint identifies file bytes; completeness and tradability require separate checks.

Low-turnover Factor

Historical exploration

Sample dates
2019-05-31 to 2026-06-30
Evidence record date
2026-09-16

60-session turnover; positive-PE common sample; monthly gross low-minus-high spread.

Revision risk and repeated historical inspection remain. After-cost performance is unverified.

Selected public evidence files

Low-valuation research

Historical exploration

Sample dates
2016-01-29 to 2026-10-16
Evidence record date
2026-10-21

Five valuation definitions; distributions and return diagnostics have separate sample periods.

Financial history and reconstructed ST status retain later-revision risk.

Selected public evidence files

Micro-cap Research

Historical exploration

Sample dates
2015-01-05 to 2026-09-03
Evidence record date
2026-10-19

Reference indices and reconstructed portfolios are separate. Dates below refer to the reconstruction snapshot.

Missing exit prices and uncalibrated real trading costs limit account-return claims.

Selected public evidence files

Cash-flow index research

Historical description

Sample dates
Periods differ across index series and experiments; see dates in each chart and table
Evidence record date
2026-09-09

Reference index history; price and total-return series are shown separately.

A complete investable reconstruction remains unverified.

Selected public evidence files

Long-run index and ETF performance

Historical description

Sample dates
Index and ETF histories have different starts and comparison windows; see each series
Evidence record date
2026-09-12

Index and ETF histories; each series retains its own dates and dividend convention.

Different periods and conventions prevent a like-for-like return ranking.

Selected public evidence files

18 Years of Style Factors

Historical exploration

Sample dates
2008-01-02 to 2026-09-04
Evidence record date
2026-09-07

Archived factor series and yearly comparisons; portfolio return is a separate question.

Historical availability and implementation remain unverified.

Selected public evidence files

Overseas signals and the A-share open

Historical exploration

Sample dates
Separate forecast and ETF starts; through 2026-10-16
Evidence record date
2026-10-18

Forecast error improvement and T+1 ETF cost scenarios are separate outcomes.

Historical delivery times are assumed; daily-bar proxies do not verify fills.

Selected public evidence files

Style and Theme Rotation

Historical exploration

Sample dates
See the period for each experiment; no common performance window
Evidence record date
2026-10-18

Different release cohorts, allocation tests and cost assumptions remain separate.

Repeated historical inspection and incomplete prospective evidence limit conclusions.

Selected public evidence files

Liquidity, Trading Costs and Capacity

Historical description

Sample dates
Common cross-market snapshot 2025-09-30
Evidence record date
2026-10-18

Trading-value distributions and cost scenarios; a shared date does not unify market coverage.

Real fills, price impact and capacity still need calibration.

Selected public evidence files
Earlier snapshot manifest and quality notes

Research topics

Derived summary

reviewed public research snapshots

Generated
2026-10-12T16:12:00+08:00
Coverage
Not provided to 2026-09-07

Microcap reconstruction

Incomplete coverage

Tushare A-share daily data with reconstructed trading rules

Generated
2026-09-12T00:00:00+08:00
Coverage
2015-01-05 to 2026-09-03
  • Membership, delistings, suspensions, and trading restrictions still require external audit.

Microcap missing-price audit

Exploratory

Tushare cleaned A-share daily data, suspension records, and special-treatment status audit

Generated
2026-09-18T17:58:00+08:00
Coverage
2015-01-05 to 2026-09-16
  • This is a missing-price classification and return-sensitivity audit only; it does not replace the strict primary NAV.
  • Of 1,848 positions with missing prices, 1,848 have suspension evidence; 210 also have special-treatment evidence. Another 274 remain unclassified. No reliable delisting-date evidence is currently available.
  • Read strict, partial, carry-all-missing, and carry-evidence-only as separate treatments.

Small-cap turnover history

Derived summary

Tushare cleaned A-share daily prices and valuation data

Generated
2026-09-18T15:35:19.472826+08:00
Coverage
2015-01-05 to 2026-09-17
  • The cleaned definition from 2015 onward is updated through the latest completed trading day, 2026-09-17.
  • The historical definition from 2008 onward is also updated through 2026-09-17, but its special-treatment and suspension eligibility fields are incomplete.

Low-turnover factor

Exploratory

Reviewed exploratory turnover snapshot

Generated
2026-09-16
Coverage
2019-05 to 2026-06
  • Independent-sample, cost, and capacity validation is not complete.

Historical style factors

Derived summary

historical market style factor result package

Generated
2026-09-07
Coverage
2008-01-02 to 2026-09-04
  • This is a historical market style reconstruction; point-in-time financial data and out-of-sample validation remain incomplete.

Cash-flow indices

Incomplete coverage

Cash-flow index recovery snapshot

Generated
2026-09-09
Coverage
Not provided to Not provided
  • Official index rules still need to be checked against the local constituent reconstruction.

Indices and ETFs

Derived summary

Index catalog and representative ETF snapshot

Generated
2026-09-12T00:00:00+08:00
Coverage
Not provided to 2026-09-07
  • Index returns are price returns. ETFs also incur fees and tracking error.

Liquidity

Derived summary

Recomputed four-market ADV20 summaries and common-date comparisons

Generated
2026-10-17T08:15:00+08:00
Coverage
2019-11-01 to 2026-10-15
  • Latest dates differ by market; the 2026 year-to-date common-date window excludes US.
  • US turnover is a close-price proxy and Japanese market caps use exploratory share history; historical point-in-time and full-universe coverage are not established.

Trading costs: static L2 study

Derived summary

Reviewed historical L2 static aggregates and event-quality audit

Generated
2026-10-12T13:18:23+08:00
Coverage
2021-01-04 to 2026-07-20
  • Ten discrete dates, not continuous history; deterministic security selection is not market-representative.
  • Full-order costs condition on ten-level coverage; dynamic impact and actual orders remain uncalibrated.
  • Reconstructed lagged daily features lack historical release-time verification; common-security comparisons condition on survival.

Trading costs: strategy diagnostic

Derived summary

Microcap strategy cost sensitivity through the shared execution simulator

Generated
2026-10-12T13:18:23+08:00
Coverage
2025-01-03 to 2025-03-31
  • 56 sessions, three capital sizes and eighteen modeled scenarios do not establish real capacity or verified net alpha.
  • Cost assumptions include later 2026 observations for counterfactual sensitivity; prices and liquidity use daily proxies.
  • Replacing costs on fixed fills creates funding shortfalls, so those NAV paths are not feasible reexecution.

Execution replay fitness and counterfactuals

Exploratory

Shenzhen event-quality review and snapshot-path counterfactuals

Generated
2026-10-12T21:12:00+08:00
Coverage
2021-01-04 to 2021-01-04
  • Full event replay remains blocked by semantics and cross-stream ordering; snapshot reconstruction reconciliation has not run.
  • Twelve paths, 10,000-share orders and five minutes exclude real queues, own impact and participation caps.
  • The missing-date-filter error is corrected; mixed-date paths are superseded by a single-day rerun.

Real-order calibration entry

Pending

Synthetic pipeline verification for the real-order entry

Generated
2026-10-12T16:12:00+08:00
Coverage
Not provided to Not provided
  • Eighty synthetic parents validate code and accounting, not real slippage; actual orders have not been connected.
  • Market-adjusted responses and chronological holdout errors do not identify causal own impact.

Liquidity data coverage and quality

Incomplete coverage

A-share published TuShare minute-to-daily reconciliation, platform coverage receipt and selected-date Level 2 aggregates

Generated
2026-10-16T22:26:30+08:00
Coverage
2022-07-15 to 2026-10-13
  • Nine exchange-calendar gaps remain and the full native minute-data semantic rebaseline is incomplete.
  • Market-wide daily totals agree closely, but 1.76% of matched security-days differ by more than 1%; the cause is unresolved.
  • Level 2 evidence covers ten selected snapshot dates; actual orders, complete event semantics and full-market coverage are absent.
  • No zero-turnover rows were observed in the recorded daily sample; this is not a whole-market zero-trading estimate.