Reversal Factor

A-share reversal factors: short-term price reversals

Reversal factors look for a bounce after a recent price move. Short holding windows create high turnover and are highly sensitive to implementation rules.

Typical direction

Recent losers are commonly ranked higher in a short-term reversal study; define the holding horizon explicitly.

Data

Adjusted price history, trading status and event filters

Refresh

One to 20 trading days; frequent rebalancing

Research hypothesis

Write the hypothesis before reading the backtest

Very short-term price deviations may partly reverse, but the signal can be overwhelmed by event risk, price limits and costs.

A factor is a testable research hypothesis, not an investment recommendation or return promise.

Factor health card

Pre-backtest checks for this factor

Research purpose

Test whether short-term deviations can be captured after realistic execution assumptions.

Refresh and rebalance

Short cycles require explicit daily or weekly execution rules.

Data timing

Use adjusted prices and a stated treatment of event days.

Neutralisation

Review sector, size and volatility exposures.

Overlapping exposures

Can conflict with momentum depending on the window.

Check before use

Test costs, slippage and limit-hit non-fills before judging the signal.

Definitions

Core measures

Short-horizon return

Adjusted price(t) ÷ adjusted price(t−N) − 1

N must match the holding rule.

Reversal spread

Recent-loser return − recent-winner return

Assess after costs and fills.

Turnover

Portfolio traded value ÷ portfolio value

Often dominates net outcomes.

Limit-hit ratio

Limit-hit names ÷ portfolio names

An A-share implementability diagnostic.

Research protocol

Keep the same research conventions across factors

Data availability

Financial, dividend and share data become available on actual disclosure or implementation dates, not report-period end dates.

Universe and exclusions

Document index membership, listing age, ST, suspensions, delistings and missing-data rules.

Processing and neutralisation

Version winsorisation, standardisation, sector/size neutralisation and missing-value rules.

Tradability

Include price limits, suspensions, participation, fees, slippage and market impact.

Out-of-sample review

Report IC, grouped returns, exposures, turnover and rolling out-of-sample evidence together.

Build and validate

What to test

  1. 1Keep formation and holding periods separate.
  2. 2Report net results and fill assumptions.
  3. 3Test event and price-limit exclusions.

Common pitfalls

  • ×Showing gross returns only.
  • ×Overlapping reversal and momentum windows without explanation.
  • ×Ignoring corporate-action and event effects.

A-share implementation

A-share checks that belong in the backtest

  • Use the actual disclosure or implementation date; do not make a field available at the report-period end date.
  • State the universe, listing-age, ST, suspension, delisting and missing-data rules before running the backtest.
  • Model price limits, suspensions, fees, slippage and participation limits instead of assuming every close can be traded.
  • Price limits and one-word boards can prevent intended reversal trades from being executed.

Research prompt

A reviewable starting prompt

In a liquid A-share universe, test five-day reversal with daily execution rules, price-limit non-fills, participation caps and costs. Report gross and net spreads, turnover, fill rates and out-of-sample stability.

FAQ

Why is turnover central to reversal?

Short holding periods cause frequent trades. A gross signal can disappear after fees, slippage and non-fills.

Can reversal and momentum be combined?

Yes, but only with clearly separated horizons and a correlation review. Otherwise the signals can offset or duplicate each other.