Aurum Wealth Management Blog


Did Everyone Fail Statistics Class?

If you graduate with a business degree from John Carroll University, there is a very high probability that you took a statistics class from Dr. Andrew Welki. The exams were grueling but he was a fair teacher, and not afraid to give you a hard time just for fun or if he expected more. During my time studying R-squared and statistical significance my sophomore year, I did not think I would use the foundation of those lessons on a daily basis like I do today.

So it makes me wonder, did all of these portfolio managers that use a static allocation strategy with only passive indices fail statistics class? Because with the evidenced based data in front of me, Dr. Welki would surely give them an ‘F’ on the final exam.

Let’s explain.

For a study to have statistical validity and ‘prove’ something, it must have an absolute t-stat ratio greater than 2. This means that the variable has a 95% chance of not being random, or having significance. So a group of professors sought to beat out a higher hurdle rate using an absolute t-stat of 3 (the 99% significance level) and tested 316 financial factors that were ‘discovered’ since 1964.

The wild chart below shows that many factors have significance, but only a few have t-stats of 5, which means they have over a 99.9% chance of being significant and less than 0.1% chance of being random. The two factors that are investable with a t-stat above 5 are circled in red: value (represented by HML) and momentum (represented by MOM).

Source: Harvey, Liu, Zhu

Talking about factors and t-stats sounds esoteric, so let’s examine historical returns to see how it worked.

Sources: AQR, Kenneth French Data Library (excess returns over cash on y-axis)

The value factor worked quite well for stock selection in the U.S. over the last 64 years. The cheapest 20% of stocks outperformed the most expensive 20% of stocks by 9% annualized. Why does the value factor exist?  Behavioral biases that are inherent within investors are likely the root cause. Investors extrapolate recent growth rates of high flying ‘glamour’ stocks, causing them to become overpriced and allowing for lower priced value stocks to earn a premium return.

Sources: AQR, Kenneth French Data Library (excess returns over cash on y-axis)

The momentum factor for U.S. stock selection showed a difference of 10% between the top quintile and bottom quintile over the past 64 years. A stock outperforming its peers over the past 2-11 months tends to continue to outperform in the short term. Anchoring bias to past views likely causes an under-reaction to news on stocks, allowing for the momentum factor to persist.

The great attribute about adding these factors to portfolios is that value and momentum, with each one’s very different fundamental reasons for existence, actually exhibit a negative correlation to one another when implemented across asset classes. Thus, including these factors in portfolios results in great diversification benefits with an overall lower risk profile and higher expected returns.

Even the kings of passive investing acknowledge the existence and persistence of factors. Vanguard, the second largest asset manager in the world, just penned a white paper on its findings even though the firm does not implement this research within its indexed products due to the active risk and cyclical variation in performance.

Asset flows today are going to passive ETFs and indices across stocks, bonds, and other asset categories. According to Morningstar’s April 2015 Fund Flows Report, over the past 1-year period, $490 billion went into passive funds across asset classes while $30 billion was sold from active funds. What passive funds do have going for them is that they are the cheapest in terms of expense ratio. However, the main factor that these passive indices look to exploit is company size, which does not show a tendency for excess returns (outperformance), although it at times has a correlation with momentum.

Perhaps all of these investors and money managers did not fail statistics class, many just might need a refresher since passive indices outperformed nearly all strategies the last few years. The evidence is clear that inclusion of factors makes a lot of common and statistical sense, providing a compliment to passive or other active allocation strategies.

For more information, please contact Aurum Wealth Management Group at 440-605-1900 or visit our website at To stay up to date with the latest investment-related news, follow us on Twitter @aurumwealth and sign up to receive our free newsletter.

  1. Campbell Harvey, Yan Liu, Heging Zhu., “…and the Cross-Section of Expected Returns.” SSRN, April 20, 2015.
  2. Scott Pappas, Joel Dickson. “Factor-based Investing.” Vanguard, April 2015.
  3. Alina Lamey. “U.S. Asset Flows Update.” Morningstar Direct, April 2015 Report, published May 18, 2015.

This material is based on public information as of the specified date, and may be stale thereafter. Aurum Wealth Management Group and/or Aurum Advisory Services has no obligation to provide updated information on the securities or information mentioned herein. Actual events may differ from those assumed and changes to any assumptions may have a material impact on any projections or estimates.

This entry was posted in Aurum Wealth Management. Bookmark the permalink. Follow any comments here with the RSS feed for this post. Comments are closed, but you can leave a trackback: Trackback URL.
© Copyright 2017 Skoda Minotti | Privacy Policy | Disclaimer | Remote Support
Cleveland 440-449-6800 | Akron 330-668-1100 | Tampa 813-288-8826
Website designed and developed by Skoda Minotti Strategic Marketing