This paper by Ryan Sullivan, Allan Timmermann, and Halbert White evaluates the performance of simple technical trading rules using White's Reality Check bootstrap methodology. The authors aim to quantify the data-snooping bias and adjust for its effect across the full universe of technical trading rules. They expand on the study by Brock, Lakonishok, and LeBaron (1992) by considering 26 trading rules applied to 100 years of daily data on the Dow Jones Industrial Average (DJIA). The paper addresses the issue of data-snooping, which can lead to spurious results, and proposes a novel procedure to correct for it. The authors construct a universe of nearly 8,000 parameterizations of trading rules and apply the bootstrap methodology to evaluate their performance. They find that certain trading rules outperform the benchmark even after adjusting for data-snooping, but the probability that the best-performing rule did not outperform the benchmark during the out-of-sample period (1987-1996) is nearly 12%, suggesting little evidence of economic value. The paper also examines the impact of transaction costs and short-sale constraints, finding no evidence that the trading rules outperform the benchmark when these factors are considered. The authors conclude that the current findings do not support the robustness of technical trading rules as previously suggested by earlier studies.This paper by Ryan Sullivan, Allan Timmermann, and Halbert White evaluates the performance of simple technical trading rules using White's Reality Check bootstrap methodology. The authors aim to quantify the data-snooping bias and adjust for its effect across the full universe of technical trading rules. They expand on the study by Brock, Lakonishok, and LeBaron (1992) by considering 26 trading rules applied to 100 years of daily data on the Dow Jones Industrial Average (DJIA). The paper addresses the issue of data-snooping, which can lead to spurious results, and proposes a novel procedure to correct for it. The authors construct a universe of nearly 8,000 parameterizations of trading rules and apply the bootstrap methodology to evaluate their performance. They find that certain trading rules outperform the benchmark even after adjusting for data-snooping, but the probability that the best-performing rule did not outperform the benchmark during the out-of-sample period (1987-1996) is nearly 12%, suggesting little evidence of economic value. The paper also examines the impact of transaction costs and short-sale constraints, finding no evidence that the trading rules outperform the benchmark when these factors are considered. The authors conclude that the current findings do not support the robustness of technical trading rules as previously suggested by earlier studies.