Press Reports

24 March 2016

Bailey and Borwein have authored numerous articles for The Conversation and the Huffington Post. Here are links to some recent articles: Bailey and Borwein have also been featured in numerous press reports. Here are links to some recent items of interest (most recent listed first):
  1. 24 Mar 2016: The CXO Advisory Group, which reviews performance of stock market "gurus," published a synopsis of the article Stock Portfolio Design and Backtest Overfitting, authored by Bailey, Jonathan Borwein and Marcos Lopez de Prado. Summarizing our article, CXO concludes, "evidence from intensive backtesting of a moderately large universe of stocks suggests that researchers can model very specific (and aggressive) performance targets in?sample, but out-?of-?sample results will exhibit extremely poor tracking and are often ruinous."
  2. 25 Dec 2015: In a Forbes article, Brett Steenbarger notes the all-too-human tendency for investors (amateur and professional) are too prone to think that they have a corner on a profitable system, when in fact they do not. He cites our blog High noon for 2015 market prophets, which highlighted the dismal record of stock pickers, hedge funds and doomsayers in 2015.
  3. 29 Sep 2015: In a Forbes article, Nathan Vardi presents an extended look at several top-ranked quantitative hedge funds, which in most cases are headed by accomplished mathematicians. In a discussion of the dangers of statistical overfitting in mathematical market models, Vardi quotes Bailey saying, "It's really hard to ascertain whether this kind of strategy has real merit or whether it's a mirage of statistical computer might." Vardi also mentions our paper on backtest overfitting, which warned that investment strategies are often "spuriously validated."
  4. 06 Aug 2015: In an Quanta interview with Neil Sloane, the curator of the Online Encyclopedia of Integer Sequences (OEIS), Sloane mentions that Jonathan Borwein looked at the odd behavior of Gregory's series for pi when truncated to 500,000 terms (or half of any power of ten), and, by consulting the OEIS, was able to identify the discrepancy terms.
  5. 20 May 2015: The Supercomputing Conference announced that the paper The NAS Parallel Benchmarks: Summary and preliminary results, which was published by Bailey and 12 co-authors in 1991, will receive the SC15 Test of Time Award. This was mentioned in several press reports, including Reuters, Business Wire and Yahoo finance. The official SC15 announcement is available here.
  6. 01 Apr 2015: In a Wall Street Journal article, Bradley Hope mentioned our paper Pseudo-mathematics and financial charlatanism. He quoted us in saying that unsound mathematical and statistical methods are "a large part of the reason why so many algorithmic and systematic hedge funds do not live up to the elevated expectations generated by their managers."
  7. 19 Apr 2015: A Buttonwood column in the Economist featured a new paper by Harvey and Liu on the prevalence of misleading statistical methods in evaluating financial strategies. This article cites our Pseudo-mathematics and financial charlatanism and The Probability of backtest overfitting in their analysis.
  8. 27 Jan 2015: Automated Trader, in an article Fiddling with figures discussed our paper Pseudo-mathematics and financial charlatanism at length, with comments by several analysts. Bailey was quoted saying, "when a statistical test is applied multiple times, the probability of obtaining false positives increases as a function of the number of trials."
  9. 04 Dec 2014: Shawn Langlois, a MarketWatch.com columnist, quoted from our article How have 2014 market prophets fared in his column This 'hideously expensive' market has given the 'show horse' pundits fits.
  10. 03 Dec 2014: Brett Steenbarger, author of several books and a news column on market trading, mentioned our article How have 2014 market prophets fared in his article Do market prophets bring market profits? Steenbarger lamented that "the performance of funds has largely fallen short of benchmarks; the doomsayers have brought doom to their followers; and even those forecasters who beat the market over a prior ten year period are likely to underperform the following year."
  11. 30 Oct 2014: Michael P. Regan, columnist for Bloomberg News, commented on our research in his article Beware Overfitting Models Even if They Win Baseball Bets. He mentioned our paper Statistical Overfitting and Backtest Performance and our new online tool demonstrating backtest overfitting.
  12. 28 Jun 2014: Jason Zweig, columnist for the Wall Street Journal, commented our our research in his article Huge Returns at Low Risk? Not So Fast. He first mentioned two humorous examples of investment strategies that would have been hugely successful if implemented over the past 20 years or so, but only by statistical accident. Then in reference to a recent claim of a strategy that has a "100% ... probability of outperformance," he quotes one of us as responding, tongue in check, "Popes have been trying to achieve infallibility for 2000 years, and these people have finally done it."
  13. 10 May 2014: John Rekenthaler, Vice President of Research for Morningstar, presented a synopsis of our "Pseudo-Mathematics" paper. He relayed our recommendation that it would "behoove the investment community to adopt a similar policy" to one now being promoted in the pharmaceutical industry, namely to make all test results public.
  14. 28 Apr 2014: The present authors' paper was mentioned in a Pacific Standard article by Ryan Jacobs. It quotes Bailey saying, "What you end up doing is that the models that you derive or you select tend to just focus on idiosyncrasies of the data, and don't have any real fundamental forward predictive power."
  15. 23 Apr 2014: After the talk by Bailey and Marcos Lopez de Prado at the "Battle of the Quants" meeting in New York City in March (see below), Bailey was invited to do a video interview for Institutional Investor Journals. This video is now available online Here.
  16. 17 Apr 2014: The present authors were featured in a Barron's article by Brendon Conway. We were quoted as saying, "The higher the number of configurations tried, the greater is the probability that the backtest is overfit."
  17. 16 Apr 2014: The present authors were featured in a Financial Times article by Stephen Foley. He summarizes our paper by saying, "The authors' argument is that, by failing to apply mathematical rigour to their methods, many purveyors of quantitative investment strategies are, deliberately or negligently, misleading clients." If you cannot view the article due the pay wall, a PDF copy is available Here.
  18. 11 Apr 2014: The present authors were featured in a Bloomberg News article. It quotes us as saying, "We strongly suspect that such back-test overfitting is a large part of the reason why so many algorithmic or systematic hedge funds do not live up to the elevated expectations generated by their managers."
  19. 10 Apr 2014: The present authors' paper "Pseudo-Mathematics and Financial Charlatanism: The Effects of Backtest Overfitting on Out-of-Sample Performance" was published by the American Mathematical Society. It is available free of charge from the AMS website, or from the SSRN website (preprint version).
  20. 16 Jan 2014: An article by Michael Oliver Weinberg at HedgeFundIntelligence.com mentions this site with the comment "we have some more intelligent colleagues with PhDs in mathematics who for sport run a website that exposes forecasters who make erroneous statistical assumptions and representations, and we tremble at the thought of falling into their cross hairs."
  21. 25 Apr 2013: Bailey was quoted in the Institutional Investor and also in the San Jose Mercury News on the potential of using high-performance computer technology to warn of impending stock market crashes.
  22. 4 Mar 2013: Bailey was quoted in Quanta (the Simons Foundation news column) and in a Wired news report on the increasingly complexity of mathematical and scientific computation, and the need for reproducibility in the field.
For additional news, see the Math Drudge news column.