Objective research to aid investing decisions
Menu
Value Allocations for September 2019 (Final)
Cash TLT LQD SPY
Momentum Allocations for September 2019 (Final)
1st ETF 2nd ETF 3rd ETF

A Few Notes on Quantitative Strategies for Achieving Alpha

Posted in Fundamental Valuation, Momentum Investing, Technical Trading

In his 2009 book, Quantitative Strategies for Achieving Alpha, flagged by Jeff Partlow, author Richard Tortoriello “seeks to determine empirically the major fundamental and market-based drivers of future stock market returns” by testing over 1,200 alternative investment strategies. He believes “that the quantitative approaches outlined in this book can provide a proven way to generate investment ideas for the qualitative investor as well as a discipline that can help improve investment results.” Richard Tortoriello is an equity research analyst with Standard & Poor’s. The principal elements of the book are:

Chapter 2: This chapter outlines the methodology applied in the book, as follows:

  • The sample period is 1987-2006.
  • The sample is about 2,200 public companies per year, trimmed in various ways to avoid “bad behavior,” such as that exhibited by microcaps and very low-priced stocks, with attention to survivorship bias and look-ahead bias.
  • The equal-weighted portfolio rebalancing period is 12 months.
  • Sorting into quintiles (five equal partitions) along the parameter(s) of interest is central to the analyses.
  • Ignore trading frictions.

A strong strategy is one for which: (1) The top (bottom) quintile significantly outperforms (underperforms) the overall sample; (2) the progression of excess returns across quintiles is systematic; (3) results are reasonably consistent over time during the sample period; and, (4) return volatility and maximum loss are reasonably low. There is no guarantee that strong strategies will continue to be strong in the future.

Chapter 3: Earnings growth and free cash flow growth are the key fundamental drivers of stock market returns, with the market efficiently incorporating the former but not the latter. The price-to-forward earnings estimate ratio is the strongest sentiment-related driver of stock market returns.

Chapters 4-10: These seven chapters provide quantitative portfolio results based on the methodology of Chapter 2 for investment strategies based on one or two indicators from among the following categories: profitability; valuation; cash flow; growth; capital allocation; price momentum; and, red flags.

Chapters 11-13: These chapters synthesize the outputs of Chapters 4-10 by providing guidance on development of an integrated investment model. Chapter 11 asserts that:

“To be successful, the common stock investor must answer three essential questions about any potential investment, with a relatively high degree of certainty: (1) Is the business doing well? (2) Is the valuation attractive? and (3) Is the timing, in terms of the overall stock market and the individual stock supply / demand trends, right?”

Chapter 12 offers some rankings of the best previously tested single-indicator and dual-indicator strategies based on specific performance metrics. Chapter 13 offers some example stock screens.

Appendix A (B) ranks the performance of 42 single-indicators (65 dual-indicator) stock sorting strategies several ways. Appendix C provides a few statistics for 43 indicators by year during 1987-2007.

Some critiques of the book are as follows:

The book seems myopic with respect to asset classes, focusing exclusively on a subset of U.S. equities. Asset class allocation is arguably more important than stock selection for overall investing success.

The strategies presented in the book involve holding a reasonably large number of individual stocks to achieve statistical reliability. Even for annual rebalancing and especially for small investors, who can allocate only modest amounts to each stock in a diversified portfolio, trading (and shorting) frictions would likely dent the alphas reported materially.

The 1987-2006 sample period coincides with a secular disinflationary environment in the U.S. that probably favored equities in general and perhaps affected specific equity screening strategies. This environment seems unlikely to persist. More generally, as acknowledged by the author, the market may be adaptive such that strategies that work historically work less well in the future.

The data snooping bias derived from testing “over 1,200 investment strategies” is also likely material, despite the mitigations offered by multiple tests and triangulations. The book offers no corrections to screen out the luck impounded by selecting the best-performing strategies from such a large pool. This bias leads to overstated alphas. The pre-emptive statement in the book that “almost all of the tests we undertook are based on existing financial and investment theory” is weak because of the generally modest and often unstable predictive power of hypotheses derived from “financial and investment theory.”

Like many comparable books and studies, this book implicitly accepts the normality (or at least non-wildness) of stock return distributions. There is a reasonable body of evidence that stock returns may be wild, such that mean returns are unstable and volatility statistics such as standard deviation and Sharpe ratio lose their “normal” meanings.

There is an argument to be made that investors experiencing investments in real time, because of uncertainties in expectations, rationally perceive anomalies differently from those hypothesizing in hindsight.

In summary, Quantitative Strategies for Achieving Alpha offers an interesting quantified overview of the performance of various fundamental, sentiment and technical indicators with respect to U.S. stock sorts over the past generation. However, investors should probably assume that the results materially overstate the size of these indicator alphas.

Why not subscribe to our premium content?
It costs less than a single trading commission. Learn more here.
Daily Email Updates
Login
Research Categories
Recent Research
Popular Posts