FoxAlgo → Research → Academic Audit → Reversal in Post-Earnings Announcement Drift

Academic audit

Failedequity event

Reversal in Post-Earnings Announcement Drift

The three-gate gauntlet · genuine only if it clears all three and survives adversarial refutation
Gate 1
Survivorship-free
free
clean panel
Gate 2
Placebo ≥ P95
P95.8
outranked ~192 of 200 baskets
Eliminated here
Gate 3
Cost-aware net
RF -0.19
net-negative after costs
Failed
Worst 12-month leg (RF)-0.90
−1.00 floor0
Every strategy here — winners included — loses in its worst 12 months. Depth is honest context, not the verdict.
Rejected at the cost gate — the net edge turns negative once modelled costs are applied.

This is an equity-event factor that trades a short-horizon reversal inside the post-earnings-announcement window, ranking stocks and taking a two-day round-trip position around the earnings drift.

What we found

The gross ranking signal has real skill: it sits at the 95.8th percentile of survivorship-free random baskets, so the raw sorting is not noise. But the effect lives on a two-day round-trip that trades often, and once realistic modelled costs are applied the net result turns negative. The risk-adjusted RF is -0.19 and the worst year's RF is -0.9. The honest reason it fails is cost: the turnover of the short holding period eats the gross rank-skill, leaving no net effect to trade.

How we tested it
2005–2026 test windowmodelled liquidity-aware costssurvivorship free
  • Data: survivorship-free 1077-name US common-stock panel, 2005-2026. Realistic modelled costs.
  • Placebo / robustness test: real result vs random baskets or shuffled signals (real vs the 95th percentile of random)
Source: Reversal within PEAD (QuantPedia)
Read the paper ↗
← The Academic Audit — all 54 studies

Research, not investment advice. “Validated” factor-legs are market-neutral diversifying building blocks with a losing worst year — none is a standalone tradeable strategy. Metrics are cost-aware and modelled (not live fills); the 2005–2026 test window is out-of-sample versus the source paper. Dollar figures are not returns and are omitted by design.