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.

Correlations

Hundreds of factors, a dozen or so real ideas.

A recurring finding in the replication literature is that the factor zoo has far less variety than its size suggests. Many “new” factors are small variations on an old one, and once you put their return streams side by side the redundancy is obvious. Below is the correlation matrix of the 18 monthly long-short series over their common window, Jul 1963–Jul 2026.

Value (B/M)
Value (E/P)
Value (CF/P)
Momentum
ST reversal
LT reversal
Profitability
Investment
Accruals
Net issuance
Low beta
Idio. vol
Total vol
Size
Div. yield
Industry mom.
Quality (QMJ)
BAB (AQR)
Value (B/M)
.7
.7
−.2
.0
.5
.1
.7
.1
.2
.3
.3
.3
.0
.7
−.1
−.0
.3
Value (E/P)
.7
.9
−.1
.0
.4
.2
.5
.1
.1
.2
.2
.3
.2
.6
−.0
.0
.3
Value (CF/P)
.7
.9
−.1
.0
.4
.2
.5
.1
.1
.2
.1
.2
.2
.5
−.0
−.0
.3
Momentum
−.2
−.1
−.1
−.3
−.2
.1
−.0
.1
.1
.2
.2
.1
−.1
−.2
.3
.2
.2
ST reversal
.0
.0
.0
−.3
.1
−.0
−.2
−.0
−.2
−.2
−.2
−.2
.2
−.0
−.1
−.2
−.0
LT reversal
.5
.4
.4
−.2
.1
−.3
.4
−.1
−.1
−.1
−.2
−.2
.4
.4
−.0
−.3
.0
Profitability
.1
.2
.2
.1
−.0
−.3
.0
.0
.4
.4
.6
.6
−.3
.1
−.1
.7
.3
Investment
.7
.5
.5
−.0
−.2
.4
.0
.1
.4
.4
.4
.4
−.1
.6
.0
.1
.3
Accruals
.1
.1
.1
.1
−.0
−.1
.0
.1
.1
.2
.1
.1
−.2
.1
.1
.1
.1
Net issuance
.2
.1
.1
.1
−.2
−.1
.4
.4
.1
.6
.6
.6
−.5
.3
−.0
.6
.2
Low beta
.3
.2
.2
.2
−.2
−.1
.4
.4
.2
.6
.8
.8
−.6
.5
.1
.6
.4
Idio. vol
.3
.2
.1
.2
−.2
−.2
.6
.4
.1
.6
.8
.9
−.7
.4
.1
.7
.3
Total vol
.3
.3
.2
.1
−.2
−.2
.6
.4
.1
.6
.8
.9
−.6
.5
.1
.7
.4
Size
.0
.2
.2
−.1
.2
.4
−.3
−.1
−.2
−.5
−.6
−.7
−.6
−.1
−.1
−.5
.0
Div. yield
.7
.6
.5
−.2
−.0
.4
.1
.6
.1
.3
.5
.4
.5
−.1
−.1
.1
.3
Industry mom.
−.1
−.0
−.0
.3
−.1
−.0
−.1
.0
.1
−.0
.1
.1
.1
−.1
−.1
.1
−.1
Quality (QMJ)
−.0
.0
−.0
.2
−.2
−.3
.7
.1
.1
.6
.6
.7
.7
−.5
.1
.1
.2
BAB (AQR)
.3
.3
.3
.2
−.0
.0
.3
.3
.1
.2
.4
.3
.4
.0
.3
−.1
.2
−1 0 +1

Most redundant

Different names, nearly the same bet.

Idiosyncratic volatility × Total volatility+0.95
Value (earnings-to-price) × Value (cash-flow-to-price)+0.88
Low beta / betting-against-beta × Total volatility+0.82
Low beta / betting-against-beta × Idiosyncratic volatility+0.79

Most opposed

Genuine diversifiers — they hedge each other.

Idiosyncratic volatility × Size (small-minus-big)-0.65
Total volatility × Size (small-minus-big)-0.59
Low beta / betting-against-beta × Size (small-minus-big)-0.55
Net share issuance × Size (small-minus-big)-0.49

The practical reading: if you already hold value, adding a second value ratio does almost nothing for you. The combinations that actually diversify are the ones in the right-hand column — most famously value and momentum, which have hedged each other for a century. The composite builder lets you see how much a spread of low-correlation signals improves the Sharpe ratio over any one of them alone.