Profits, prophets and pseudoscience

Market prophets

We have all seen articles on financial news sites confidently predicting the future of various markets, often with remarkable specificity. Here are some that have appeared just in the past few weeks (as of 5 November 2018):

  1. The stock market is setting up for another rally, according to Elliott Wave theory.
  2. Market strategist urges aggressive buying, says bull market could last decades.
  3. Short-seller who warned of ‘unavoidable pain’ earlier this year turns bullish.
  4. Stock strategist who saw market correction now predicts 10% to 14% rally — and a fizzle in 2019.
  5. Why the Dow Jones Industrial Average should be higher in six months.
  6. Stocks could rally 20% after this bruising rout, says Guggenheim’s Minerd — after that, watch out.
  7. The Wall Street analyst who called this stock-market rout sees another nasty drop for the S&P 500.
  8. A ‘lost decade’ for stocks may have just arrived, says this adviser.
  9. Charting an ominous technical tilt, S&P 500 plunges from the 200-day average.
  10. Would you be prepared if the Dow Jones Industrial Average were to fall 5,700 points?.
  11. Get ready for an 8% to 13% stock market correction.

Do predictions work?

A moment’s reflection shows that predictions such as the above and many others cannot possibly be taken as a reliable foretelling of financial markets. After all, if the person making such a prediction truly had a thoroughly researched, carefully vetted and actionable method for predicting a financial market, he or she most certainly would not be disclosing the strategy or specific predictions in any public forum, much less to a widely read financial news venue. Instead, this person would be hard at work making appropriate investments in the market, as discreetly as possibly so as to avoid trades being undermined by competitors, front-runners and arbitragers.

What’s more, given today’s high-tech markets, it is not clear that any prediction based on a relatively unsophisticated analysis of economic trends or market data could possibly be better than dart throwing. Keep in mind that most market trading nowadays is dominated by highly mathematical algorithms and big-data machine learning operations, where even the slightest statistically significant trading opportunity is rapidly arbitraged away. The net result of all this high-tech trading is that markets are reduced to a random walk, one characteristic of which is utter unpredictability.

Does technical analysis work?

Many of the predictions above, and others like them, are based on technical analysis, or more broadly on human analysis of charts, graphs, “waves” and trends. Indeed, it is hard to survey any financial news source without seeing at least one or two articles of this genre. Many investors, both individual and institutional, use technical analysis methods in decision making.

But do these methods really work? Do they really deliver consistently above-market-index results? In a word, “no.” As market analyst Laszlo Birinyi, interviewed in the book The Heretics of Finance, bluntly wrote, “The truth is technical analysis doesn’t work in the market.”

The failure of technical analysis is discussed in greater detail in an earlier Mathematical Investor blog.

Scientific evaluations of market forecasts

Recently Nir Kaissar analyzed a set of predictions by professional market forecasters over a 17-year period from 1999 through 2016. He found that although there was a reasonably high correlation between the average forecast and the year-end price of the S&P 500 index for the given year, these predictions were surprisingly unreliable during major shifts in the market. For example, Kaissar found that the strategists overestimated the S&P 500’s year-end price by 26.2 percent on average during the three recession years 2000 through 2002, yet they underestimated the index’s level by 10.6 percent for the initial recovery year 2003. A similar phenomenon was seen in 2008, when strategists overestimated the S&P 500’s year-end level by a whopping 64.3 percent in 2008, but then underestimated the index by 10.9 percent for the first half of 2009. In other words, as Kaissar lamented, the forecasts were least useful when they mattered most.

In 2017 one of the present authors and his colleagues published an in-depth analysis of 68 market forecasters, including many prominent figures in the industry, some of whom employ technical analysis and others of whom do not. Expanding on an earlier study by the CXO Advisory Group, they analyzed forecasts based on two key factors:

  • The time frame of the forecast. Forecasts are categorized as up to one month, up to three months, up to nine months or beyond nine months.
  • The importance and specificity of the forecast. For example, a forecast that states “the market will be volatile in the next few days” is not a very specific forecast, but the forecast “the market will experience a correction” is more specific and thus more important.

The study found that the average accuracy score of these forecasts was 48%, not significantly different than chance, and the distribution of scores followed a Gaussian bell curve indistinguishable from that of a distribution of random deviates. In short, there was no statistically significant evidence of any overall skill.

Are market forecasts helpful?

In light of evidence such as this, many financial analysts have concluded that market forecasts are of little use in setting a rational long-term investment strategy. James Mackintosh, writing in the Wall Street Journal, lamented that most market predictions for 2017 turned out “pathetically wrong,” yet few were willing to admit the nearly universal failure. Eve Kaplan, writing in Forbes, conceded that “most of these forecasts have little predictive power.”

As mentioned above, some market forecasts are based on assessments of economic data. But as Miles Johnson of the Financial Times writes, it is unclear whether economic data-driven predictions are any better than others. Other forecasts are based on “reversion to the mean” principles. But again, as Johnson notes, these predictions have been laughably inaccurate in recent years.

Follow the money

Why indeed is the investing world, individual and institutional, so keen on “knowing” the future, or even thinking that it is possible to “know” the future? At the very least, it should be clear that market prediction is very difficult, well beyond the realm of conventional efforts by human forecasters — there are far too many variables and uncertainties in extremely complex systems such as the financial markets for one to have any confidence in such a forecast. And if someone could consistently predict any market, even for a short period of time, with even slightly better than chance rates of success, such a person would not be operating a blog, writing a financial news column or appearing on a television show, but instead would be using this critical skill to reap millions in profits.

Part of the problem, sadly, is monetary. Online financial news venues are desperate for clicks, cable news channels are desperate for viewers, and print news sources are desperate for subscribers, and so all have a financial incentive to present sensational, eye-catching material. And, regrettably, the public has an unhealthy fascination with sensational news, particularly financial predictions and doomsday scenarios. By comparison, carefully researched, modestly worded and rigorously peer-reviewed scientific studies are seldom of interest to the public and do not generate many ad clicks on commercial sites.

But there is more at stake here than advertising revenue or even financial performance. Professionals in the finance and investment field need to carefully consider the long-term impact of pseudoscientific market predictions on the credibility of the field in the public eye. As a colleague of one of the present authors observes,

It is unfathomable that [the public] would be exposed to nonsense on such a continuous basis. It is tantamount to the Journal of the American Medical Society routinely publishing the pronouncements of a shaman, voodoo practitioner or bloodletting practitioner.

One of the present authors and his colleagues expressed a similar sentiment:

Historically, scientists have led the way in exposing those who utilize pseudoscience to extract a commercial benefit. As early as the eighteenth century, physicists exposed the nonsense of astrologers. Yet mathematicians in the twenty-first century have remained disappointingly silent with regard to those in the investment community who, knowingly or not, misuse mathematical techniques such as probability theory, statistics, and stochastic calculus. Our silence is consent, making us accomplices in these abuses.

Indeed, silence is consent. To the extent that those of us in the mathematical financial field continue to wink our eyes at headline-grabbing pseudoscientific market predictions, we are complicit in this fraud.

Comments are closed.