Literature Review

Brown and Jennings (1989) showed that technical analysis has value in a model in which prices are not fully revealing and traders have rational conjectures about the relation between prices and signals. Frankel and Froot (1990) showed evidence for the rising importance of chartists. Neftci (1991) showed that a few of the rules used in technical analysis generate well-defined techniques of forecasting, but even well-defined rules were shown to be useless in prediction if the economic time series is Gaussian. However, if the processes under consideration are nonlinear, then the rules might capture some information. Tests showed that this may indeed be the case for the moving average rule. Taylor and Allen (1992) report the results of a survey among chief foreign exchange dealers based in London in November 1988 and found that at least 90 per cent of respondents placed some weight on technical analysis, and that there was a skew towards using technical, rather than fundamental, analysis at shorter time horizons. In a comprehensive and influential study Brock, Lakonishok and LeBaron (1992) analysed 26 technical trading rules using 90 years of daily stock prices from the Dow Jones Industrial Average up to 1987 and found that they all outperformed the market. Blume, Easley and O’Hara (1994) show that volume provides information on information quality that cannot be deduced from the price. They also show that traders who use information contained in market statistics do better than traders who do not. Neely (1997) explains and reviews technical analysis in the foreign exchange market. Neely, Weller and Dittmar (1997) use genetic programming to find technical trading rules in foreign exchange markets. The rules generated economically significant out-of-sample excess returns for each of six exchange rates, over the period 1981–1995.

in the absence of trading costs. yet. it is perhaps naïve to work on the premise that ‘bull’ and ‘bear’ markets exist. Neely and Weller (2001) use genetic programming to show that technical trading rules can be profitable during US foreign exchange intervention. Fern´andez-Rodr´ıguez. Gonz´alez-Martel and Sosvilla-Rivero (2000) apply an artificial neural network to the Madrid Stock Market and find that. exchange rate predictability is dramatically reduced. technical analysis was more popular at shorter time horizons. Cesari and Cremonini (2003) make an extensive simulation comparison of popular dynamic strategies of asset allocation and find that technical analysis only performs well in Pacific markets.Lui and Mole (1998) report 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. . again. the technical trading rule is always superior to a buyand-hold strategy for both ‘bear’ market and ‘stable’ market episodes. Secondly. several technical indicators do provide incremental information and may have some practical value. Mamaysky and Wang (2000) examines the effectiveness of technical analysis on US stocks from 1962 to 1996 and finds that over the 31-year sample period. Lo. Lee and Swaminathan (2000) demonstrate the importance of past trading volume. when using technical analysis in the foreign exchange market. LeBaron (1999) shows that. They found that over 85% of respondents rely on both methods and. longterm. Neely (1998) reconciles the fact that using technical trading rules to trade against US intervention in foreign exchange markets can be profitable. after removing periods in which the Federal Reserve is active. One criticism I have is that beating the market in the absence of costs seems of little significance unless one is interested in finding a signal which will later be incorporated into a full system. the intervention tends to be profitable. but not in a ‘bull’ market.

Kavajecz and Odders-White (2004) show that support and resistance levels coincide with peaks in depth on the limit order book 1 and moving average forecasts reveal information about the relative position of depth on the book.Cheol-Ho Park and Scott H. an excellent review paper on technical analysis. for example. While technical analysts today may employ trading rules based. Irwin wrote ‘The profitability of technical analysis: A review’ Park and Irwin (2004). Literature review Technical analysis claims the ability to forecast the future direction of asset prices through the study of past market data. According to Nison (1991.g. asserts that the stock market moves in certain phases with predictable patterns. . developed by Charles Dow and re¯ned by William Peter Hamilton in the 1800s.. While the classic book Murphy (1986) summarizes the the Dow Theory and various other technical indicators. 13). They also show that these relationships stem from technical rules locating depth already in place on the limit order book. such as the relative strength index. and whose techniques evolved into what is known today as the candlestick patterns. there is a growing and large literature on new techniques of technical analysis due to the wide availability of data and computing power (see. Covel (2005) and Kirkpatrick and Dahlquist (2006)). the Dow Theory. p. In the United States. on various price transformations and other market statistics. e. among the ¯rst and famous technicians (who use past prices to predict future price movements) is the legendary speculator Munehisa Homma who amassed a huge fortune in the rice market in the 1700s in Japan.

and so the use of the random walk model rules out any value of technical analysis. for which Schwert (2003) provides an excellent survey. There is a huge literature on stock predictability recent examples of which are Ferson and Harvey (1991). and Ang and Bekaert (2006). supporting studies. in the earlier years by many. With robust statistical tests. Lakonishok. The market e±ciency was interpreted.cycles and momentum oscillators. when Fama and Blume (1966) showed that common ¯lter rules are not pro¯table based on daily prices of 30 individual securities in the Dow Jones Industrial Average (DJIA) over 1956{1962. Goyal and Welch (2003). they ¯nd that simple trading rules. Lo and MacKinlay (1988) provide a variance ratio speci¯cation test that completely rejects the random walk model. The predictability of stock returns allows for the possibility of pro¯table technical rules. These empirical ¯ndings have perhaps prompted Fama (1970) to propose the well known e±cient market hypothesis that market prices re°ect all available information so that no abnormal returns can be made with historical price and other market data. the stock return must be predictable. as a random walk model for the stock price. that various economic variables can forecast stock returns. and LeBaron (1992) provide strong evidence on the pro¯tability of technical trading. based on the . also suggest predictable patterns of the stock returns. Lo and MacKinlay (1999). such as Fama and Schwert (1977) and Campbell (1987). Cowles (1933) seems to be the ¯rst to conduct an empirical study of technical analysis that is published in an academic journal. who ¯nds that Hamilton's forecasts based on the Dow Theory over the period of 1904 and 1929 are successful only 55% of the time. For any technical trading rule to be pro¯table. the moving averages (MAs) are the most popular and simple rules. various asset pricing anomalies. Brock. In addition. such as Campbell and Thompson (2007) and Cochrane (2007). Similar conclusion is also reached by Jensen and Benington (1970) in their study of relative strength systems. Subsequent studies on technical analysis are few until in the 1960s. Indeed. However. Current studies. provide further evidence even on out-of-sample predictability.

Moreover. show that there are substantial gains with the use of MAs and the gains are much larger than those in the stock market.g. Moreover. Lo. it is di±cult to show the true e®ectiveness of technical trading rules because of a data-snooping bias (see. though. Lo and MacKinlay. and Wang (2000) also ¯nd that technical analysis has added value to the investment process based on their novel approach comparing the distribution conditional on technical patterns. For example. rules . In its simplest form.. with the unconditional distribution. outperforms the market over the 90 year period prior and up to 1987 based on daily data on DJIA. Gehrig and Menkho®c (2006) argue that technical analysis today is as important as fundamental analysis to currency mangers. e. the results in foreign exchange markets are generally much stronger. Mamaysky. in their comprehensive study of applying both kernel estimators and automated rules to hundreds of individual stocks.popular MAs and range breakout. In contrast to the equity markets. among others. such as head-and-shoulders and double-bottoms. 1990). Statistically. LeBaron (1999) and Neely (2002). which occurs when a set of data is used more than once for the purpose of inference and model selection.

In a two-period model with third period consumption. In a model of information asymmetry. Brown and Jennings (1989) show that rational investors can gain from forming expectations based on historical prices.4 However. Grundy and Kim (2002) also ¯nd value of using technical analysis. Sullivan. Lakonishok. and White (1999) show via bootstrap that Brock. One could then argue that a bootstrap is subject to speci¯cation bias and that generic algorithms can be inadequate due to ine±cient ways of learning. In an equilibrium model where the volume also plays a role. there are no theoretical studies closely tied to the conventional use of .that are invented and tested by using the same data set are likely to exaggerate their e®ectiveness. Timmermann. Allen and Karjalainen (1999) ¯nd little pro¯tability in technical trading. for example. Theoretically. Blume. it appears that the statistical debate on the e®ectiveness of technical analysis is unlikely to get settled soon. Accounting for the data-snooping bias. few studies explain why technical analysis has value under certain conditions. and O'Hara (1994) show that traders who use information contained in market statistics do better than traders who do not. Easley. to our knowledge. In any case. Using generic algorithms. and LeBaron's results are much weakened.

can be optimally used to add value to the investment process. we study the classic asset allocation problem and examine how technical analysis. especially the MA. In so doing. The exploratory study here attempts to ¯ll this gap of the literature. .technical analysis. nor are there studies that calibrate the model to data to provide insights on the realistic use of technical analysis in practice.

Master your semester with Scribd & The New York Times

Special offer for students: Only $4.99/month.

Master your semester with Scribd & The New York Times

Cancel anytime.