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
9 topics across 4 research areas

Current findings by topic

Choose a question, read the finding and its limits, then explore the evidence. Each topic keeps its own sample dates.

01

Stock characteristics

Valuation and turnover: how stock groups differ.

Historical exploration

Low-turnover Factor

Low-turnover stocks performed better in this historical comparison. The gap became smaller after accounting for other stock characteristics.

Sample dates
2019-05-31 to 2026-06-30
Main unresolved question

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

Read the study →
Methods and research historyFull finding

In this historical sample, the low-turnover group averaged 1.38 percentage points more per month than the high-turnover group. After accounting for the full listed controls and regrouping on the adjusted score, the gap was about 1.20 points. These are group comparisons. A consistent trading test is still needed to establish after-cost portfolio performance.

What is compared
60-session turnover; positive-PE common sample; monthly gross low-minus-high spread.
Historical exploration

Low-valuation research

Different accounting definitions change valuation comparisons. Full-market coverage and achievable returns remain unverified.

Sample dates
2016-01-29 to 2026-10-16
Main unresolved question

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

Read the study →
Methods and research historyFull finding

Valuation comparisons depend on accounting definitions and usable stocks. Full-market coverage and achievable trading returns remain unverified.

What is compared
Five valuation definitions; distributions and return diagnostics have separate sample periods.
02

Portfolios and indices

Micro-caps, cash flow and index histories, with their reconstruction limits.

Historical exploration

Micro-cap Research

Reconstructed small-cap portfolios suffered deep drawdowns. Missing exit prices complicate the return calculation.

Sample dates
2015-01-05 to 2026-09-03
Main unresolved question

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

Read the study →
Methods and research historyFull finding

Rebuilt portfolios of the smallest stocks show large losses and long waits to recover old highs. Some prices and delisting outcomes are missing, so these historical results do not yet show what a real account could have earned.

What is compared
Reference indices and reconstructed portfolios are separate. Dates below refer to the reconstruction snapshot.
Historical description

Cash-flow index research

Compare cash-flow index histories, drawdowns and rule reconstructions while keeping official indices and shadow portfolios separate.

Sample dates
Periods differ across index series and experiments; see dates in each chart and table
Main unresolved question

A complete investable reconstruction remains unverified.

Read the study →
Methods and research historyFull finding

Past cash-flow index results can be studied, but a real stock portfolio that follows those rules has not yet been fully rebuilt and checked. Price-only returns and returns including dividends answer different questions.

What is compared
Reference index history; price and total-return series are shown separately.
Historical description

Long-run index and ETF performance

Read index and ETF histories with their dates and dividend conventions. Different series cannot be ranked on equal terms.

Sample dates
Index and ETF histories have different starts and comparison windows; see each series
Main unresolved question

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

Read the study →
Methods and research historyFull finding

The index and ETF charts show different historical market paths. Their dates, fees, and treatment of dividends differ, so the numbers are not directly comparable investment results.

What is compared
Index and ETF histories; each series retains its own dates and dividend convention.
03

Market styles and timing

Long-run styles, overseas information and theme rotation.

Historical exploration

18 Years of Style Factors

Value, size and other styles behaved differently across years and market regimes. Historical factor results need separate trading validation.

Sample dates
2008-01-02 to 2026-09-04
Main unresolved question

Historical availability and implementation remain unverified.

Read the study →
Methods and research historyFull finding

Nineteen archived factor series differ widely and overlap. Their history and tradability still need verification.

What is compared
Archived factor series and yearly comparisons; portfolio return is a separate question.
Historical exploration

Overseas signals and the A-share open

Separate overseas-market forecasting information from the return of a T+1 ETF trading scenario.

Sample dates
Separate forecast and ETF starts; through 2026-10-16
Main unresolved question

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

Read the study →
Methods and research historyFull finding

Opening forecasts improve mainly through the US market factor. After FX and SPX controls, the Asian basket adds little stable information.

What is compared
Forecast error improvement and T+1 ETF cost scenarios are separate outcomes.
Historical exploration

Style and Theme Rotation

Published experiments have not established a stable trading advantage from theme-to-ETF signals.

Sample dates
See the period for each experiment; no common performance window
Main unresolved question

Repeated historical inspection and incomplete prospective evidence limit conclusions.

Read the study →
Methods and research historyFull finding

The published experiments do not establish a stable theme-to-ETF trading advantage.

What is compared
Different release cohorts, allocation tests and cost assumptions remain separate.
04

Trading and liquidity

Costs, liquidity and the gap between a simulation and real fills.

Historical description

Liquidity, Trading Costs and Capacity

Compare trading activity and cost scenarios before judging strategy capacity. Real fills and price impact still need calibration.

Sample dates
Common cross-market snapshot 2025-09-30
Main unresolved question

Real fills, price impact and capacity still need calibration.

Read the study →
Methods and research historyFull finding

Trading-value comparisons show how much the sampled stocks traded and how complete the data is. They do not yet tell us how large an order could be filled or what trading would cost.

What is compared
Trading-value distributions and cost scenarios; a shared date does not unify market coverage.
About the latest editorial updateReading update · 22 October 2026

Low-turnover and valuation reports clarify their evidence limits. This editorial update adds no new return observations.