Anomaly Replication Lab
  • The zoo
  • Screener
  • Multiple testing
  • Correlations
  • Composite
  • Methodology
  • Guides
  • Resources
Sign in
Anomaly Replication LabSign in
  • The zoo
  • Screener
  • Multiple testing
  • Correlations
  • Composite
  • Methodology
  • Guides
  • Resources

Educational replication, not advice. Long-short returns are decile spreads or published factor series from the Kenneth R. French Data Library and AQR, value-weighted, US equities, and gross of transaction and shorting costs — which for high-turnover signals like short-term reversal is the whole story. Reproducing a premium in-sample is not evidence it will persist. Past performance does not predict future results.

© 2026 Austin Levenberg. All rights reserved. This site’s code, design, written analysis, and computed results are not licensed for reuse or redistribution — terms of use · privacy.

Data: Kenneth R. French Data Library, AQR Capital Management public data sets.

Part of austinlev.dev — four research tools, and a note on where this is headed.

← the zoo
MomentumHolds up

Industry momentum

Buy the industries that have gone up, short the ones that have gone down.

Moskowitz & Grinblatt · Journal of Finance · published 1999 · original sample 1963–1995 (reported ≈ +8.0%/yr, t ≈ 3.5)

How our version differs from the paper. Constructed here from the 30 French industry portfolios (top third minus bottom third on prior 12-1 return), not from a paper file. The decay figure compares our replication in-sample to our replication after publication, so it isn’t affected by this — but the level gap between “reported” and “our replication” partly is.
Our replication
+6.5%
1927–2026, per year
Full-sample t
6.1
Sharpe 1.25
Post-publication
+5.4%
since 2000, t = 2.5
Decay
31%
of the premium is gone

The whole history

Growth of $1 in the long-short strategy, gross of trading and shorting costs. The dashed line marks the year the paper was published — everything to its right is out-of-sample.

$1 in the long-short strategy, gross of costs. Grey = the sample the paper had; blue = everything since.

Before and after publication

The comparison the site is built around.

WindowReturn/yrVolSharpet-statMax DD
Paper's sample (1963–1995)+7.9%4.2%1.835.3-12%
After publication (2000–2026)+5.3%5.6%0.962.5-35%
Full sample (1927–2026)+6.5%5.1%1.256.1-51%

Against the factor models

Is there a risk-adjusted alpha left once you control for the standard factors? Alpha is annualized; t-stats use Newey–West standard errors.

ModelAlpha/yrAlpha tR²Biggest loadings
CAPM (market only)+7.1%5.80.01Mkt-RF -0.04
Fama–French 3-factor+7.3%6.10.02HML -0.04, Mkt-RF -0.04, SMB -0.02
Fama–French 5-factor+7.4%6.10.04HML -0.06, RMW -0.06, CMA +0.06
FF5 + momentum+6.5%5.70.11Mom +0.10, RMW -0.07, SMB -0.03

Rolling performance

Trailing 3-year Sharpe ratio. Long stretches below zero are common.

By decade

1920s+23.8%/yrSharpe 6.99
1930s-1.7%/yrSharpe -0.20
1940s+5.6%/yrSharpe 1.45
1950s+7.3%/yrSharpe 3.12
1960s+10.7%/yrSharpe 3.00
1970s+8.8%/yrSharpe 1.88
1980s+6.3%/yrSharpe 1.47
1990s+7.2%/yrSharpe 1.91
2000s+3.6%/yrSharpe 0.54
2010s+4.1%/yrSharpe 0.88
2020s+9.9%/yrSharpe 1.96

The risk explanation

Industries have common cash-flow shocks; industry momentum may be time-varying industry risk.

The behavioral explanation

Sector-rotation herding and slow diffusion of industry-wide news.

Why it might have decayed

Moskowitz–Grinblatt claimed industry momentum explained most of individual-stock momentum; Grundy–Martin (2001) pushed back hard. As a standalone it has been weak since the late 1990s.

Long-short = top-third minus bottom-third of 30 industry portfolios ranked on prior 12-1 month return. Value-weighted, US equities, monthly, gross of transaction and shorting costs. The original-paper figures are approximate and annualized for comparison.