Информационный центр "Центральный Дом Знаний"
Онлайн всего: 1 Гостей: 1 Пользователей: 0
|
Brock W., Lakonishok J., LeBaron B. Simple Technical Trading Rules and the Stochastic Properties ofBrock W., Lakonishok J., LeBaron B. Simple Technical Trading Rules and the Stochastic Properties of Stock Returns WILLIAM BROCK, JOSEF LAKONISHOK, and BLAKE LeBARON* ABSTRACT 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, our 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, 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. The term "technical analysis" is a general heading for a myriad of trading techniques. Technical analysts attempt to forecast prices by the study of past prices and a few other related summary statistics about security trading. They believe that shifts in supply and demand can be detected in charts of market action. Technical analysis is considered by many to be, the original form of investment analysis, dating back to the 1800s. It came into widespread use before the period of extensive and fully disclosed financial information, which in turn enabled the practice of fundamental analysis to develop. In the United States, the use of trading rules to detect patterns in stock prices is probably as old as the stock market itself. The oldest technique is attributed to Charles Dow and is traced to the late i800s. Many of the techniques used today have been utilized for over 60 years. These techniques for discovering hidden relations in stock returns can range from extremely simple to quite elaborate. * Brock and LeBaron are from the University of Wisconsin and Lakonishok is from the University of Illinois. We are grateful to Tim Bollerslev, К. C. Chan, Louis Chan, Eugene Fama, Bruce Lehmann, Mark Ready, Jay Ritter, William Schwert, Theo Vermaelen, and the editor, Rene Stulz, and anonymous referees. The paper was presented at the AFA Meetings at New Orleans, The Amsterdam Institute of Finance, the NBER Summer Institute, the University of Limburg, and The Wharton School. We are thankful to Hank Pruden, a leading technical analyst, for guidance on some of the technical analysis literature. Brock was partially supported by the National Science Foundation (SES 87-20671), the Vilas Trust, and the University of Wisconsin Graduate School. LeBaron was partially supported by the National Science Foundation (SES 91-09671), and the University of Wisconsin Graduate School. The attitude of academics towards technical analysis, until recently, is well described by Malkiel (1981): Obviously, I am biased against the chartist. This is not only a personal predilection, but a professional one as well. Technical analysis is anathema to the academic world. We love to pick on it. Our bullying tactics are prompted by two considerations: (1) the method is patently false; and (2) it's easy to pick on. And while it may seem a bit unfair to pick on such a sorry target, just remember: it is your money we are trying to save. Nonetheless, technical analysis has been enjoying a renaissance on Wall Street. All major brokerage firms publish technical commentary on the market and individual securities, and many of the newsletters published by various "experts" are based on technical analysis. In recent years the efficient market hypothesis has come under serious siege. Various papers suggested that stock returns are not fully explained by common risk measures.1 A line of research directly related to this work provides evidence of predictability of equity returns from past returns.2 In general, the results of these studies are in sharp contrast with most earlier studies that supported the random walk hypothesis- and concluded that the predictable variation in equity returns was economically and statistically very small. Two competing explanations for the presence of predictable variation in stock returns have been suggested: (1) market inefficiency in which prices take swings from their fundamental values, and (2) markets are efficient and the predictable variation can be explained by time-varying equilibrium returns.3 There is no evidence so far that unambiguously distinguishes these two competing hypotheses. Although many earlier studies concluded that technical analysis is useless, the recent studies on predictability of equity returns from past returns suggest that this conclusion might have been premature.4 In this paper we explore two of the simplest and most popular technical rules: moving average-oscillator and trading-range break (resistance and support levels). In the first method, buy and sell signals are generated by two moving averages, a long period, and a short period. In the second method signals are generated as stock prices hit new highs and lows. These rules will be evaluated by their ability to forecast future price changes. For statistical inferences, standard tests will be augmented with the bootstrap methodology inspired by Efron (1979), Freedman and Peters (1984a, 1984b), and Efron and Tibshirani (1986). Following this methodology, returns from an artificial Dow series are generated and the trading rules are applied to the series. Comparisons are then made between returns from these simulated series and the actual Dow Jones series. Neither the bootstrap methodology nor the use of technical analysis to evaluate model specifications are in particular new to the finance literature. The contribution of this paper lies in the combination of these two techniques. This procedure allows testing a wide range of null models. When models are rejected by such a statistical test, information is provided on how to modify the model to achieve a better description of the series. In addition, the trading rules used in this paper may have power against certain alternatives that are difficult to detect using standard statistical tests. Few, if any, empirical tests in financial economics are free of the data-instigated pre-test biases discussed in Learner (1978).5 The more scrutiny a collection of data receives, the more likely "interesting" spurious patterns will be observed. Stock prices are probably the most studied financial series and, therefore, most susceptible to data snooping. In addition, Merton (1987) suggests that individuals have a tendency to come up with "exciting" spurious results (anomalies): All this fits well with what the cognitive psychologists tell us is our natural individual predilection to focus, often disproportionately so, on the unusual... This focus, both individually and institutionally, together with little control over the number of tests performed, creates a fertile environment for both unintended selection bias and for attaching greater significance to otherwise unbiased estimates than is justified. Therefore, the possibility that various spurious patterns were uncovered by technical analysis cannot be dismissed. Although a complete remedy for data-snooping biases does not exist, we mitigate this problem: (1) by reporting results from all our trading strategies, (2) by utilizing a very long data series, the Dow Jones index from 1897 to 1986, and (3) emphasizing the robustness of results across various nonoverlapping subperiods for statistical inference.6 Our study reveals that technical analysis helps to predict stock price changes. The patterns uncovered by technical rules cannot be explained by first order autocorrelation and by changing expected returns caused by changes in volatility. To put it differently, the trading profits are not consistent with a random walk, an AR(1), a GARCH-M model, or an Exponential GARCH. The results generally show that returns during buy periods are larger than returns during sell periods. Moreover, returns during buy periods are less volatile than returns during sell periods. For example, the variable-length moving-average strategy produced on average a daily return for buy periods of 0.042 percent, which is about 12 percent per year. In contrast, the corresponding daily return for the sell periods is -0.025 percent, or about — 7 percent per year. This strategy results in a daily standard deviation of 0.89 percent for buy periods and a higher one, 1.34 percent, for sell periods. The remainder of the paper is organized as follows: Section I describes the data and our technical trading rules; Section II presents the empirical results of the tests utilizing traditional techniques; Section III describes the bootstrap methodology, Section IV presents the empirical results from the bootstrap simulations, and Section V concludes and summarizes our results. |
Loading
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||
Simple Technical Trading Rules and the Stochastic Properties of Stock Returns (1992)