Objective research to aid investing decisions

Value Investing Strategy (Strategy Overview)

Allocations for December 2022 (Final)
Cash TLT LQD SPY

Momentum Investing Strategy (Strategy Overview)

Allocations for December 2022 (Final)
1st ETF 2nd ETF 3rd ETF

Stock Anomaly Momentum Strategy

| | Posted in: Momentum Investing

Do U.S. stock return anomalies exhibit exploitable momentum? In their December 2016 paper entitled “Scaling Up Market Anomalies”, Doron Avramov, Si Cheng, Amnon Schreiber and Koby Shemer test momentum across stock return anomalies. Their investment universe consists of the long and short sides of 15 stock portfolios, each long (short) the top (bottom) tenth of stocks based on sorting by one of the following 15 variables: failure probability, O-Score, net stock issuance, composite equity issuance, total accruals, net operating assets, momentum, gross profitability, asset growth, return on assets, abnormal capital investment, standardized unexpected earnings, analyst dispersion, idiosyncratic volatility and book-to-market ratio. They each month rank the 15 anomaly portfolios by prior-month return and test an anomaly momentum strategy that is long (short) the long (short) sides of the top five winner (bottom five loser) portfolios. They also consider top three-bottom three and top four-bottom four long-short strategies. Their benchmark is the equally weighted combination of all 15 anomaly portfolios. Using daily and monthly data for a broad sample of U.S. common stocks during 1976 through 2013, they find that:

  • Returns of individual anomaly portfolios are volatile and deteriorate over time. Investor sentiment, market illiquidity and past market return are strong predictors of anomaly performance. Investor sentiment is the best predictor of anomaly performance since 2000.
  • Pairwise correlations of returns among these portfolios are low, and the equally weighted benchmark therefore greatly mitigates downside risk. For this benchmark:
    • Gross average monthly return is 0.63% over the entire sample period, with gross monthly Sharpe ratio 0.10 and gross monthly three-factor (market, size, book-to-market) alpha 0.80%.
    • Gross monthly Sharpe ratio is 0.08 (0.14) during 1976-1999 (2000-2013).
  • Individual anomaly portfolio returns have meaningful persistence (strong positive autocorrelations), indicating that anomalies exhibit momentum. More directly, over the entire sample period:
    • The long top five-short bottom five strategy has gross average monthly return 1.08%, with gross monthly Sharpe ratio 0.18 and gross monthly three-factor alpha 1.27%.
    • The long top four-short bottom four strategy has gross average monthly return 1.19%, with gross monthly Sharpe ratio 0.19 and gross monthly three-factor alpha 1.38%.
    • The long top three-short bottom three strategy has gross average monthly return 1.28%, with gross monthly Sharpe ratio 0.19 and gross monthly three-factor alpha 1.47%.
  • However, gross performances of the anomaly momentum strategies are weaker during 2000-2013 than during 1976-1999.
  • Anomaly momentum strategies conditioned on investor sentiment perform well during 2000-2010.

In summary, evidence indicates that U.S. stock return anomalies exhibit potentially exploitable monthly momentum.

Cautions regarding findings include:

  • Performance results are gross, not net. Trading frictions from monthly portfolio reformation and shorting costs may be substantial. Moreover, shorting may not be feasible for all stocks selected for strategy short sides.
  • The 15 selected anomalies are already known to perform well during the sample period. These anomalies may impound data snooping bias from prior research, thereby overstating expectations.
  • Testing three anomaly momentum strategies and (particularly) 11 variations of each on the same data impounds additional snooping bias, such that the best results overstate expected performance.
  • As noted in the paper, values for the selected investor sentiment measure end in 2010, so associated tests are shorter and stale.
  • The complex data collection/processing requirements are beyond the reach of most investors. Delegating these tasks to fund managers would involve administrative and management fees.

See also “Beat the Market with Hot-Anomaly Switching?”.

Login
Daily Email Updates
Filter Research
  • Research Categories (select one or more)