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
Low riskGone

Idiosyncratic volatility

Stocks with high stock-specific volatility earn LOW returns — the 'IVOL puzzle'.

Ang, Hodrick, Xing & Zhang · Journal of Finance · published 2006 · original sample 1963–2000 (reported ≈ +12.0%/yr, t ≈ 3.7)

How our version differs from the paper. AHXZ's effect is concentrated in the short leg of expensive, hard-to-short small caps; the value-weighted spread is weaker and noisier. 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
+2.6%
1963–2026, per year
Full-sample t
1.7
Sharpe 0.23
Post-publication
+0.6%
since 2007, t = 0.1
Decay
94%
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 (1963–2000)+6.4%23.4%0.382.1-76%
After publication (2007–2026)-3.3%27.3%0.020.1-79%
Full sample (1963–2026)+2.6%25.9%0.231.7-80%

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)+12.4%4.20.28Mkt-RF -0.89
Fama–French 3-factor+10.8%4.90.62SMB -1.41, HML +0.63, Mkt-RF -0.54
Fama–French 5-factor+5.1%2.80.74RMW +1.22, SMB -1.11, CMA +0.61
FF5 + momentum+3.4%1.80.75RMW +1.20, SMB -1.11, CMA +0.56

Rolling performance

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

By decade

1960s-7.1%/yrSharpe -0.35
1970s+6.0%/yrSharpe 0.27
1980s+21.3%/yrSharpe 1.31
1990s+3.5%/yrSharpe 0.15
2000s+11.5%/yrSharpe 0.30
2010s+4.2%/yrSharpe 0.25
2020s-6.3%/yrSharpe -0.18

The risk explanation

Textbook theory says idiosyncratic risk shouldn't be priced at all — so a negative price is a genuine puzzle, not a premium.

The behavioral explanation

High-IVOL stocks are lottery-like and hard to arbitrage (costly to short, hard to value), so overpricing persists.

Why it might have decayed

The effect is concentrated in the very short leg — expensive, hard-to-short small caps — and much of it overlaps the low-beta and profitability stories. Weaker since the mid-2000s.

Long-short = top-decile minus bottom-decile of stocks univariately sorted on the characteristic. Value-weighted, US equities, monthly, gross of transaction and shorting costs. The original-paper figures are approximate and annualized for comparison.