No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Technical Analysis Around the World
Abstract
Over 5,000 popular technical trading rules are not consistently profitable in the 49 country indices that comprise the Morgan Stanley Capital Index once data snooping bias is accounted for. Each market has some rules that are profitable when considered in isolation but these profits are not statistically significant after data snooping bias adjustment. There is some evidence that technical trading rules perform better in emerging markets than developed markets, which is consistent with the finding of previous studies that these markets are less efficient, but this result is not strong. While we cannot rule out the possibility that these trading rules compliment other market timing techniques or that trading rules we do not test are profitable, we do show that over 5,000 trading rules do not add value beyond what may be expected by chance when used in isolation during the time period we consider.
... Technical analysis has its origins in the Dow Theory and the works of Cowles [22], Dow and Hamilton [23] [24] from early 20 th century. Since that early time, the effectiveness of technical analysis when compared to fundamental and quantitative analysis has been a matter of controversy, with references testifying for the effectiveness of technical analysis [25] [26] and against it [27]. Different researchers test variations of these strategies using different data, which makes it very difficult to compare and contrast different researches. ...
... Evidences in this paper might be getting the same results as previous studies did, in which technical "rules are profitable when considered in isolation, but these profits are not statistically significant after" adjustment for data snooping and survivorship bias [27]. ...
... In that sense, random walks and components of the S&P 500 Index might be just too efficient for momentum strategies to perform. "There is some evidence that technical trading rules perform better in emerging markets" due to their inefficiencies [27], technical strategies tend to perform better in inefficient markets [51]. ...
To promote economic stability, finance should be studied as a hard science, where scientific methods apply. When a trading strategy is proposed, the underlying model should be transparent and defined robustly to allow other researchers to understand and examine it thoroughly. Like any hard sciences, results must be repeatable to allow researchers to collaborate, and build upon each other's results. Large-scale collaboration, when applying the steps of scientific investigation, is an efficient way to leverage "crowd science" to accelerate research in finance. In this paper, we demonstrate how a real world problem in economics, an old problem still subject to a lot of debate, can be solved by the application of a crowd-powered, collaborative scientific computational framework, fully supporting the process of investigation dictated by the modern scientific method. This paper provides a real end-to-end example of investigation to illustrate the use of the framework. We intentionally selected an example that is self-contained, complete, simple, accessible, and of constant debate in both academia and the industry: the performance of a trading strategy used commonly in technical analysis. Claims of efficiency in technical analysis, referred derisively by some sources as "Black Magic", are of widespread use in mainstream media and usually met with a lot of controversy. In this paper we show that different researchers assess this strategy differently, and the subsequent debate is due more to the lack of method than purpose. Most results reported are not repeatable by other researchers. This is not satisfactory if we intend to approach finance as a hard science. To counterweight the status quo, we demonstrate what one could do by using collaborative and investigative features of contributions and leveraging the power of crowds.
... The latest Morgan Stanley Capital International (MSCI) 2018 classifies Indonesia as an emerging market in Asia. An emerging markets are interesting because some researches show that there exist inefficiency in emerging markets [3] [4]. The existence of inefficiency gives favor to the technical analysis. ...
... They found that MA and TRB trading rules after the inclusion of transaction costs generated profits that declined after 1986. Marshall et al. (2009) applied MA and TRB technical trading rules on daily index return for 23 established and 26 developing stock markets and observed that both MA and TRB technical trading rules performed better in developing than the established stock markets. Chen et al. (2011) re-inspected the applicability and validity of technical analysis in the Taiwan stock market during the period January 1975 to December 2006, by applying the White Reality Check and the Hansen Superior Productive Ability tests to nullify the data snooping bias. ...
This paper inspects whether variable- and fixed-length moving averages (VMA and FMA), and trading range breakout (TRB) rules have prognostic capability and can earn profits superior to buy-and-hold plan, when applied on KSE-100 index of Pakistan stock market during the full sample period January 1, 1997 to December 31, 2013. Full sample results provided empirical evidence for VMA rule that it has significant predictive power and is able to generate profits superior to simple buy-and-hold plan even after inclusion of transaction costs. The highest mean buy returns yielded by VMA, FMA and TRB rules are seen in noncrises periods. The overall implication of this study is that traders in the Pakistan stock market can utilize this information to obtain excess returns on a regular basis.
... Although many studies have seemingly validated the general notion of market efficiency and the related ineffectiveness of technical trading rules (e.g., Marshall, Cahan and Cahan [2010]), even Eugene Fama, the "father of market efficiency," in an interview with Financial Engineering News [Mehtais, 2006] admitted that the one area that cannot be resolved is the issue of momentum. The seminal work in this area was conducted by Jagadeesh and Titman [1993], who find that stocks having the highest (lowest) return over the past three-to twelve-month period continue to perform well (poorly) over the subsequent period. ...
... Consistent with these results, Schulmeister (2009) and Park and Irwin (2007) find that technical trading rules only yield economic profits in U.S. markets until the late 1980s, but not thereafter. Marshall et al. (2010) conduct a test of technical analysis using data from 49 countries, including the Netherlands and the U.S., and conclude that technical analysis performs better in emerging markets than in developed ones. Smith et al. (2013) report that about one-third of actively managed equity and balanced funds use technical analysis, with the latter generating values for alpha and volatility which are generally higher. ...
─ Although in developed countries the futures markets have been in existence since mid-nineteenth century, they are relatively new in the developing countries. In the late twentieth and early twenty first century, many new futures exchanges were established in developing countries and in the majority of these newly established exchanges substantial growth in futures trading have been observed within a small period of time. This fast growth in the futures trading volume was mainly due to the tremendous leverage the futures provides to speculators. Thanks to futures margining system, by committing only a small fraction of the money needed to maintain a position on the underlying security in the spot market, a speculator can attain a much higher return potential by buying or selling a futures contract. This paper studies this effect by employing daily return data on 19 selected stocks listed continuously in IMKB-30 (Istanbul Stock Exchange 30, Turkey) from January 2005 to December 2010. Popular technical indicators are used to generate buy and sell signals in both the spot market (Istanbul Stock Exchange) and the futures market (VOB, a fast growing derivatives exchange located in İzmir, Turkey). The profit/loss resulting from trading strategies are then calculated and compared. The results of the study show that, although the amount invested in both markets is the same, the profit generated from the strategies applied on futures is significantly higher than that on spot market. A CAPM (Capital Asset Pricing Model) based hedge ratio is used to apply the trading strategies generated INTERNATIONAL JOURNAL OF ECONOMICS AND FINANCE STUDIES Vol 4, No 2, 2012 ISSN: 1309-8055 (Online) 92 from spot market data on futures. The results show that this strategy generates superior returns in the futures market.
We use a least squares metric to match the return pattern of a target stock with that of an out-of-sample-twin. The twin with the smallest metric is found by a comprehensive period-by-period search of stocks in the Center for Research in Security Prices data set extending back to 1926. If technical analysis has value, targets of twins producing the highest returns in the twin postperiod should also have the highest performance in the target postperiod. Using a randomly selected sample of 66,000 return patterns, we find higher means for targets corresponding to the highest returning twin quintile. We also use regressions to risk adjust target returns and find that twin returns in the postmatch period significantly predict risk-adjusted target returns.
Discount brokerage and online trading have revolutionized the world of individual investing, but not always for the good. One way of seeing the impact of these changes is to contrast the behavior of recent investors with their counterparts from several decades ago. In this respect, we examine three datasets to study how discount brokerage and online trading and the associated surge of financial data available online, compares to investing with full-fee brokers during the 1960s. In comparative terms, we find a marked shift from the use of fundamental analysis to the use of technical analysis and intuition. We also find a shift towards speculative activity. Overall, these shifts have been detrimental to investors, as they are associated with lower returns, increased risk, greater turnover, and more concentrated portfolios. To the best of our knowledge, we are the first to estimate the marginal contribution of reporting technical analysis as a strategy.
This paper examines the profitability of momentum strategies implemented on international stock market indices. Our results indicate statiscally significant evidence of momentum profits. The momentum profits arise mainly from time-series predictability in stock market indices very little profit comes from predictability in the currency markets. We also find higher profits for momentum portfolios implemented on markets with higher volume in the previous period, indicating that return continuation is stronger following an increase in trading volume. This result confirms the informational role of volume and its applicability in technical analysis.
This paper investigates whether stock-price indexes of seventeen emerging markets can be characterized as random walk (unit root) or mean reversion processes. We implement a test that can account for structural breaks in the underlying series and is more powerful than standard tests. We find that for fourteen countries, stock prices exhibit structural breaks. Furthermore, for ten countries, the null hypothesis of a random walk can be rejected at the one or 5% significance level. Our results indicate that ignoring structural breaks that arise from the liberalization of emerging markets can lead to incorrect inference that these indices are characterized by random walks, and are consistent with the points made by Bekaert et al. [J. Int. Money Finan. 21 (2002) 295]. Our findings hold true regardless of whether stock indexes are denominated in US dollar terms, in local currencies terms, or in real terms.
We report findings from a survey of United States foreign exchange traders. Our results indicate that: (i) in recent years electronically-brokered transactions have risen substantially, mostly at the expense of traditional brokers; (ii) the market norm is an important det e rminant of interbank bid-ask spread and the most widely-cited reason for deviating from the conventional bid-ask spread is a thin/hectic market; (iii) half or more of market respondents believe that large players dominate in the dollar-pound and dollar-Sw i ss franc markets; (iv) technical trading best characterizes about 30% of traders, with this proportion rising from five years ago; (v) news about macroeconomic variables is rapidly incorporated into exchange rates; (vi) the importance of individual macroe c onomic variables shifts over time, although interest rates always appear to be important; (vii) economic fundamentals are perceived to be more important at longer horizons, while short-run deviations from the fundamentals are attributed to excess speculat i on and institutional customer/hedge fund manipulation; (viii) speculation is generally viewed positively, as enhancing market efficiency and liquidity, even though it exacerbates volatility; (ix) central bank intervention does not appear to have substanti a l effect, although there is general agreement that it increases volatility, and finally; (x) traders do not view purchasing power parity as a useful concept, even though a significant proportion (40%) believe that it affects exchange rates at horizons of over six months.
This study examines the potential profit of ten Variable Length Moving Average (VMA) technical trading rules in ten emerging equity markets in Latin America and Asia from January 1982 through April 1995. The average difference in buy − sell returns after trading costs for each rule and country are compared to a buy and hold strategy. Taiwan, Thailand and Mexico emerge as markets where technical trading strategies may be profitable. We find no strong evidence of profitability for the other markets. However, we find that 82 out of the 100 country–trading rule combinations tested in ten emerging markets, disregarding their statistical significance, correctly predict the direction of changes in the return series. These findings may provide investors with important asset allocation information.
We report evidence on the profitability and statistical significance among 2,127 technical trading rules. The best rules are found to be significantly profitable based on standard tests. We then employ White's (2000) Reality Check to evaluate these rules and find that data-snooping biases do not change the basic conclusions for the full sample. A sub-sample analysis indicates that the data-snooping problem is more serious in the second half of the sample. Profitability becomes much weaker in the more recent period, suggesting that the foreign exchange market becomes more efficient over time. Evidence from cross exchange rates confirms the basic findings.
In this paper we examine the ability of filter rules to predict variation in expected daily returns for a sample of 120 Dow Jones and S&P 100 stocks from 1963 through 1989. Equally weighted portfolios of filter-rule-traded stocks consistently outperform a buy-and-hold portfolio of the same stocks before accounting for transaction costs. The difference in returns between filter rule and buy-and-hold portfolios is eliminated by one-way transaction costs of 12 basis points. The economic significance of daily stock return autocorrelations is estimated. A marginal 1 percent increase in a first-order autocorrelation increases filter rule returns by an estimated 3.84 percent.
This study evaluates the profitability of technical trading rules in Pacific-Basin equity markets by using equilibrium asset pricing models with time-varying expected returns. Specifically, this study uses asset pricing models under complete integration, mild segmentation [Errunza, V., Losq, E., Padmanablan, P., 1985. International asset pricing under mild segmentation hypothesis. Journal of Banking and Finance, 16, 949–972.], and complete segmentation in assessing trading rule profits. The same set of technical rules that Brock et al. [Brock, W., Lakonishok, J., LeBaron, B., 1992. Simple technical trading rules and the stochastic properties of stock returns, Journal of Finance, 47, 1731–1764.] examine are applied to the Japanese, U.S., Canadian, Indonesian, Mexican and Taiwanese equity indices. The results from the standard tests indicate that the technical rules have significant forecast power for all countries, except for the U.S. However, the results from the bootstrap tests indicate that some equilibrium asset pricing models (mainly, the asset pricing model under mild segmentation) are consistent with the observed trading rule returns for Japan, the U.S., the recent period of Canada and Taiwan. The overall results indicate that taking into account the time-varying expected returns is important to evaluate the profitability of the technical trading rules.
This article reports the results of a questionnaire survey conducted in February 1995 on the use by foreign exchange dealers in Hong Kong of fundamental and technical analyses to form their forecasts of exchange rate movements. Our findings reveal that>85% of respondents rely on both fundamental and technical analyses for predicting future rate movements at different time horizons. At shorter horizons, there exists a skew towards reliance on technical analysis as opposed to fundamental analysis, but the skew becomes steadily reversed as the length of horizon considered is extended. Technical analysis is considered slightly more useful in forecasting trends than fundamental analysis, but significantly more useful in predicting turning points. Interest rate-related news is found to be a relatively important fundamental factor in exchange rate forecasting, while moving average and/or other trend-following systems are the most useful technical technique.
Quantitative market timing strategies are not consistently profitable when applied to 15 major commodity futures series. We conduct the most comprehensive study of quantitative trading rules in this market setting to date. We consider over 7000 rules, employ two alternative bootstrapping methodologies, account for data-snooping bias, and consider different time periods. We cannot rule out the possibility that trading rules compliment some other trading strategy or that some traders may have success using a specific rule on its own, but we do conclusively show that none of these rules beat the market any more than expected given random data variation.
In this paper, we analyze the usefulness of technical analysis, specifically the widely employed moving average trading rule from an asset allocation perspective. We show that, when stock returns are predictable, technical analysis adds value to commonly used allocation rules that invest fixed proportions of wealth in stocks. When uncertainty exists about predictability, which is likely in practice, the fixed allocation rules combined with technical analysis can outperform the prior-dependent optimal learning rule when the prior is not too informative. Moreover, the technical trading rules are robust to model specification, and they tend to substantially outperform the model-based optimal trading strategies when the model governing the stock price is uncertain.
In the finance literature, statistical inferences for large-scale testing problems usually suffer from data snooping bias. In this paper we extend the “superior predictive ability” (SPA) test of Hansen (2005, JBES) to a stepwise SPA test that can identify predictive models without potential data snooping bias. It is shown analytically and by simulations that the stepwise SPA test is more powerful than the stepwise Reality Check test of Romano and Wolf (2005, Econometrica). We then apply the proposed test to examine the predictive ability of technical trading rules based on the data of growth and emerging market indices and their exchange traded funds (ETFs). It is found that technical trading rules have significant predictive power for these markets, yet such evidence weakens after the ETFs are introduced.
This paper tests two of the simplest and most popular trading rules — moving average and trading range break-out — in the Chilean stock market. Overall, our results are similar to the ones of Brock et al. (1992), providing strong support for the technical strategies. In fact, buy signals consistently generate higher returns than sell signals. Moreover, returns following sell signals are negative, which is not easily explained by any of the currently existing equilibrium models. However, we do not observe any difference regarding the risk for the signs of buys and sells, a result explained by the fact that the Chilean stock market is highly concentrated and illiquid.
We revisit the apparent historical success of technical trading rules on daily prices of the DJIA from 1897 to 2008. We use the False Discovery Rate as a new approach to data snooping. The advantage of the FDR over existing methods is that it selects more outperforming rules and di- versi es against model uncertainty. Persistence tests show that an investor would never have been able to select ex ante the future best-performing rules. Moreover, even the in-sample performance is completely oset by the introduction of transaction costs. Overall, our results seriously call into question the economic value of technical trading rules.
This study extends the work of Brock et al.’s (1992) empirical analysis on technical trading rules (price and momentum) by including trading volume moving averages; broader indices (New York Stock Exchange (NYSE) and National Association of Security Dealers Automatic Quotations (NASDAQ)) covering both large-cap and small-cap firms using market weightings; and focusing on a time period that includes great innovations in trading and disseminating data to the market. Similar to their study, we base our conclusions on nonparametric analysis. By extending the t-test analysis through a residual bootstrap methodology utilizing a random walk, a generalized autoregressive conditional heteroskedasticity in mean (GARCH-M), and a GARCH-M with instrument variables, criticisms of earlier technical analysis are mitigated. Overall, the results support Brock et al.’s (1992) price-weighted index (Dow Jones Industrial Average (DJIA)) analysis by showing that the technical trading rules add value by capturing profit opportunities when compared to a buy-and-hold strategy. When the analysis of the trading rules are applied to different time periods, the results reveal a weakening in profit potential over time. This may imply that the market is becoming more efficient in disseminating information to a wider range of investors.
Technical analysis, also known as 'charting,' has been a part of financial practice for many decades, but this discipline has not received the same level of academic scrutiny and acceptance as more traditional approaches such as fundamental analysis. One of the main obstacles is the highly subjective nature of technical analysis-the presence of geometric shapes in historical price charts is often in the eyes of the beholder. In this paper, we propose a systematic and automatic approach to technical pattern recognition using nonparametric kernel regression, and we apply this method to a large number of U.S. stocks from 1962 to 1996 to evaluate the effectiveness of technical analysis. By comparing the unconditional empirical distribution of daily stock returns to the conditional distribution-conditioned on specific technical indicators such as head-and-shoulders or double bottoms-we find that over the 31-year sample period, several technical indicators do provide incremental information and may have some practical value. Copyright The American Finance Association 2000.
The aim in this paper is to replicate and extend the analysis of visual technical patterns by Lo et al. (2000) using data on the UK market. A non-parametric smoother is used to model a nonlinear trend in stock price series. Technical patterns, such as the 'head-and-shoulders' pattern, that are characterised by a sequence of turning points are identified in the smoothed data. Statistical tests are used to determine whether returns conditioned on the technical patterns are different from random returns and, in an extension to the analysis of Lo et al. (2000), whether they can outperform a market benchmark return. For the stocks in the FTSE 100 and FTSE 250 indices over the period 1986 to 2001, we find that technical patterns occur with different frequencies to each other and in different relativities to the frequencies found in the US market. Our extended statistical testing indicates that UK stock returns are less influenced by technical patterns than was the case for US stock returns. Copyright Blackwell Publishers Ltd, 2003.
We assess whether some simple forms of technical analysis can predict stock price movement in Asian markets. We find the rules to be quite successful in the emerging markets of Malaysia, Thailand and Taiwan. The rules have less explanatory power in more developed markets such as Hong Kong and Japan. On average for our sample, mean percentage changes in stock indices on days that the rules emit buy signals exceed means on days that the rules emit sell signals by 0.095% per day, or about 26.8% on an annualized basis. We estimate “break-even” round-trip transactions costs (which would just eliminate gains from technical trading) to be 1.57% on average. We also find that technical signals emitted by U.S. markets have substantial forecast power for Asian stock returns beyond that of own-market signals. This is consistent with the reasoning that the technical rules identify periods when global equilibrium expected returns deviate substantially from their unconditional mean.
In econometric applications, often several hypothesis tests are carried out at once. The problem then becomes how to decide which hypotheses to reject, accounting for the multitude of tests. This paper suggests a stepwise multiple testing procedure that asymptotically controls the familywise error rate. Compared to related single-step methods, the procedure is more powerful and often will reject more false hypotheses. In addition, we advocate the use of studentization when feasible. Unlike some stepwise methods, the method implicitly captures the joint dependence structure of the test statistics, which results in increased ability to detect false hypotheses. The methodology is presented in the context of comparing several strategies to a common benchmark. However, our ideas can easily be extended to other contexts where multiple tests occur. Some simulation studies show the improvements of our methods over previous proposals. We also provide an application to a set of real data. Copyright The Econometric Society 2005.
Data snooping occurs when a given set of data is used more than once for purposes of inference or model selection. When such data reuse occurs, there is always the possibility that any satisfactory results obtained may simply be due to chance rather than to any merit inherent in the method yielding the results. This problem is practically unavoidable in the analysis of time-series data, as typically only a single history measuring a given phenomenon of interest is available for analysis. It is widely acknowledged by empirical researchers that data snooping is a dangerous practice to be avoided, but in fact it is endemic. The main problem has been a lack of sufficiently simple practical methods capable of assessing the potential dangers of data snooping in a given situation. Our purpose here is to provide such methods by specifying a straightforward procedure for testing the null hypothesis that the best model encountered in a specification search has no predictive superiority over a given benchmark model. This permits data snooping to be undertaken with some degree of confidence that one will not mistake results that could have been generated by chance for genuinely good results.
Technica, or chartist, analysis of financial markets involves providing forecasts or trading advice on the basis of largely visual inspection of past prices, without regard to any underlying economic or 'fundamental' analysis. This paper reports the results of a questionnaire survey, conducted on behalf of the Bank of England, among chief foreign exchange dealers based in London in November 1988. Amongst other findings, it is revealed that at least 90 per cent of respondents place some weight on this form of non-fundamental analysis when forming views at one or more time horizons. There is also a skew towards reliance on technical, as opposed to fundamentalist, analysis at shorter horizons, which becomes steadily reversed as the length of horizon considered is increased. A very high proportion of chief dealers view technical and fundamental analysis as complementary forms of analysis and a substantial proportion suggest that technical advice may be self-fulfilling. (JEL F31).
The anomalous returns associated with net stock issues, accruals, and momentum are pervasive; they show up in all size groups (micro, small, and big) in cross-section regressions, and they are also strong in sorts, at least in the extremes. The asset growth and profitability anomalies are less robust. There is an asset growth anomaly in average returns on microcaps and small stocks, but it is absent for big stocks. Among profitable firms, higher profitability tends to be associated with abnormally high returns, but there is little evidence that unprofitable firms have unusually low returns. Copyright (c) 2008 The American Finance Association.
This paper tests two of the simplest and most popular trading rules--moving average and trading range break--by utilizing the Dow Jones Index from 1897 to 1986. Standard statistical analysis is extended through the use of bootstrap techniques. Overall, their results provide strong support for the technical strategies. The returns obtained from these strategies are not consistent with four popular null models: the random walk, the AR(1), the GARCH-M, and the Exponential GARCH. Buy signals consistently generate higher returns than sell signals, and further, the returns following buy signals are less volatile than returns following sell signals. Moreover, returns following sell signals are negative, which is not easily explained by any of the currently existing equilibrium models. Copyright 1992 by American Finance Association.
We propose a new test for superior predictive ability. The new test compares favorable to the reality check for data snooping (RC), because the former is more powerful and less sensitive to poor and irrelevant alternatives. The improvements are achieved by two modifications of the RC. We employ a studentized test statistic that reduces the influence of erratic forecasts and invoke a sample dependent null distribution. The advantages of the new test are confirmed by Monte Carlo experiments and in an empirical exercise, where we compare a large number of regression-based forecasts of annual US inflation to a simple random walk forecast. The random walk forecast is found to be inferior to regression-based forecasts and, interestingly, the best sample performance is achieved by models that have a Phillips curve structure.
Numerous studies in the finance literature have investigated technical analysis to determine its validity as an investment tool. Several of these studies conclude that technical analysis does have merit, however, it is noted that the effects of data-snooping are not fully accounted for. In this paper we utilize White's Reality Check bootstrap methodology (White (1997)) to evaluate simple technical trading rules while quantifying the data-snooping bias and fully adjusting for its effect in the context of the full universe from which the trading rules were drawn. Hence, for the first time, the paper presents a means of calculating a comprehensive test of performance across all trading rules. In particular, we consider the study of Brock, Lakonishok, and LeBaron (1992), expand their universe of 26 trading rules, apply the rules to 100 years of daily data on the Dow Jones Industrial Average, and determine the effects of data-snooping. During the sample period inspected by Brock, Lakonishok and LeBaron, we find that the best technical trading rule is capable of generating superior performance even after accounting for data- snooping. However, we also find that the best technical trading rule does not provide superior performance when used to trade in the subsequent 10-year post-sample period.
Stepwise multiple testing as formalized data snooping Data-snooping, technical trading rule performance, and the bootstrap
- J Romano
- M Wolf
Romano, J. and Wolf, M., 2005, Stepwise multiple testing as formalized data snooping, Econometrica 73, 1237-1282. r24 Sullivan, R., A. Timmermann, and H. White, 1999, Data-snooping, technical trading rule performance, and the bootstrap, Journal of Finance 24(5), 1647-1691
Technical trading-rule profitability, data snooping, and reality check: Evidence from the foreign exchange market
- P-H Hsu
- C-M Kuan
Hsu, P-H. and C-M. Kuan, 2005. Technical trading-rule profitability, data snooping, and
reality check: Evidence from the foreign exchange market, Journal Financial Econometrics
3(4), 606-628.