A reader asked: “This paper tests 5,000 technical trading rules and comes up empty on all of them. Not even say 200 were good. Has this paper been vetted by anyone? Is it iron-clad? The authors write: ‘We find that many technical trading rules produce statistically significant profits before consideration is given to data snooping bias, but this profitability disappears after data snooping bias is taken into account.’ How are they doing ‘data snooping’ detection?”
The paper you cite is an update of the version summarized at “Technical Analysis Tested Globally”. Working papers such as this one have varying degrees of vetting (peer review). The authors do not note the degree of vetting in this paper. The corresponding author (named on the first page) would likely be amenable to inquiries.
Some related research is at:
“Technical Analysis Tested on Long-run DJIA Data”
“Does Technical Trading Work with Commodity Futures?”
“Technical Analysis: ‘Anathema to the Academic World’?”
Summaries of Chapters 8 and 9 in “Evidence-Based Technical Analysis: Applying the Scientific Method and Statistical Inference to Trading Signals (Chapter-by-Chapter Review)”
“Does Technical Trading Work for Certain Kinds of Stocks?”
See the notes on Chapter 6 in “Evidence-Based Technical Analysis: Applying the Scientific Method and Statistical Inference to Trading Signals (Chapter-by-Chapter Review)” for some qualitative aspects of data snooping bias. The book itself presents the mathematics of correcting for data snooping bias.
See also “The Demon’s Drudgery” for additional discussion.
