Please rotate your device to landscape mode to view the charts.

Anomalies and the Expected Market Return

Journal: Journal of Finance

Date: 20220201

Author: Dong, Xi; Li, Yan; Rapach, David E.; Zhou, Guofu

Abstract:
We provide the first systematic evidence on the link between long-short anomaly portfolio returns--a cornerstone of the cross-sectional literature--and the time-series predictability of the aggregate market excess return. Using 100 representative anomalies from the literature, we employ a variety of shrinkage techniques (including machine learning, forecast combination, and dimension reduction) to efficiently extract predictive signals in a high-dimensional setting. We find that long-short anomaly portfolio returns evince statistically and economically significant out-of-sample predictive ability for the market excess return. The predictive ability of anomaly portfolio returns appears to stem from asymmetric limits of arbitrage and overpricing correction persistence.

Link: Google Scholar

Background and Context

Research Focus

The study examines whether long-short anomaly portfolio returns from cross-sectional research can predict aggregate market excess returns, linking two major areas of finance research.

Data

Analysis uses 100 representative stock market anomalies from 1970-2017, examining their ability to forecast market returns through various statistical techniques.

Methodology

Employs machine learning, forecast combination, and dimension reduction to efficiently extract predictive signals from anomaly returns while avoiding overfitting.

Superior Performance of Anomaly-Based Market Return Forecasts

  • Shows the predictive power of different forecasting methods using anomaly returns
  • All methods outperform the traditional benchmark (exceeding 0.5% threshold)
  • C-ENet achieves the highest R² of 2.81%, indicating strongest predictive ability

Asymmetric Predictive Power Across Market Segments

  • Demonstrates stronger predictive ability in short-leg returns compared to long-leg returns
  • Long-short combined strategy provides the best predictive performance
  • Highlights asymmetric limits of arbitrage between long and short positions

Enhanced Prediction During High Market Friction Periods

  • Shows stronger predictive power during periods of high market friction
  • Short fees and liquidity constraints have largest impact on prediction accuracy
  • Demonstrates how market frictions affect the speed of mispricing correction

Impact on Arbitrageurs' Trading Positions

  • Shows how anomaly returns predict changes in arbitrageurs' positions
  • Negative values indicate reduction in net market positions following anomaly signals
  • Demonstrates real-world trading impact of anomaly-based predictions

Economic Value of Anomaly-Based Predictions

  • Illustrates economic gains from using anomaly-based prediction strategies
  • PLS strategy achieves highest utility gain of 6.38% annually
  • All strategies show economically significant benefits for investors

Contribution and Implications

  • First systematic evidence linking cross-sectional anomalies to market return predictability
  • Demonstrates practical value of combining multiple anomaly signals for market prediction
  • Provides new insights into how market frictions affect price discovery and return predictability

Data Sources

  • R² statistics visualization based on Table II of the article
  • Portfolio components analysis based on Table II's long-leg and short-leg results
  • Market friction effects based on Table IV data
  • Arbitrage trading impact based on Table V coefficients
  • Economic value visualization based on utility gain statistics reported in Figure 2