تجزیه و تحلیل فنی و سرمایه گذاران فردی

We find that individual investors who use technical analysis and trade options frequently make poor portfolio decisions, resulting in dramatically lower returns than other investors. The data on which this claim is based consists of transaction records and matched survey responses of a sample of Dutch discount brokerage clients for the period 2000–2006. Overall, our results indicate that individual investors who report using technical analysis are disproportionately prone to have speculation on short-term stock-market developments as their primary investment objective, hold more concentrated portfolios which they turn over at a higher rate, are less inclined to bet on reversals, choose risk exposures featuring a higher ratio of nonsystematic risk to total risk, engage in more options trading, and earn lower returns.

The intersection between the literature on individual investors and the literature on technical analysis is sparse. As a result, knowledge about individual investors’ use of technical analysis has been limited. In the present paper, we present the results of a new study which deepens our understanding about how using technical analysis impacts individual investors’ portfolios. The existing literature on technical analysis effectively ignores the experience of individual investors. Instead, it emphasizes its efficacy, the time periods in which its use is associated with abnormal trading profits, and the markets where such abnormal profits have been earned.1 In a comment about individual investors that serves as an exception, Neely (1997) describes the profitable use of technical analysis to trade in foreign exchange markets, but states the following: “Technical trading is much less useful for individuals, who would face much higher transactions costs and must consider the opportunity cost of the time necessary to become an expert on foreign exchange speculating and to keep up with the market on a daily basis… In addition to higher transactions costs, individual investors following technical rules also must accept the risk that such a strategy entails.” (p. 31). We note that Neely provides no empirical evidence to support his remarks about individual investors’ use of technical analysis. Most of what is known about the actual use of technical analysis by individual investors comes from a study of U.S. investors by Lewellen et al. (1980) (LLS),2 and is based on transaction records and matched survey responses from the period 1964–1970. In line with Neely's remark, the findings from this study suggest that technical analysis severely degrades the performance of individual investors’ portfolios. LLS report that investors who trade the most frequently use technical analysis to a disproportionate degree and underperform other investors by 4.1% per year on a risk-adjusted basis. This result is economically important: LLS find that 27% of the investors in their sample use technical analysis. The LLS results about technical analysis being both costly to individual investors and widespread were economically important during the 1960s. But are these results robust in respect to time and space? This is a critical question, and forms the starting point of our investigation. The existing literature on individual investor behavior since LLS has effectively ignored technical analysis. We believe this is because of a major difference in the type of data LLS used in their study and the data used in more recent studies by other authors. LLS combine their transaction data with matched survey data in which investors report the strategies they use (such as technical analysis), the investment objectives they maintain (such as achieving short-term capital gains), and other related information. In contrast, the data used by more recent studies, such as those by Odean, 1998a and Odean, 1999 and Barber and Odean, 2000 and Barber and Odean, 2008), only include account-level transactions, not survey information.3 In this paper, we report that the key findings from LLS (1980) are robust to time and space, and moreover are driven by investors’ decisions about portfolio concentration, turnover, and options trading. Our study uses data from the Netherlands which cover the period 2000–2006 and are from a discount broker where investors trade online. These data consist of transaction records and matched survey responses, the same structure used by LLS.4 LLS apply the term “high roller” to describe high-turnover investors, and associate high rollers with the use of technical analysis as a strategy and achieving short-term capital gains as an objective. Notably, LLS make no effort to isolate the separate effects of technical analysis, a focus on achieving short-term capital gains, and high turnover. In contrast, we do, and believe that ours is the first paper to isolate the impact of individual investors’ use of technical analysis on concentration, turnover, derivatives use, betting on reversals, risk-taking, and returns. Our new findings constitute the major contributions of the current paper. We find that investors who report using technical analysis hold more concentrated portfolios than other investors, and have higher ratios of nonsystematic risk to total risk. They also trade more frequently than other investors, especially in respect to options. As a result of these behavior patterns, investors using technical analysis earn lower raw and risk-adjusted returns than other investors. The magnitudes are economically important: controlling for concentration and turnover, the marginal cost associated with technical analysis is approximately 50 basis points of raw return per month. Turnover associated with technical analysis adds a further 20 basis points per month of cost. Concentration adds an additional 2 basis points. A major finding from our study concerns investors who both trade options frequently and use technical analysis. For “high derivative rollers,” the marginal cost of technical analysis from poor portfolio selection is 140 basis points, not the 50 basis points which we find for the full sample of investors, with turnover linked to technical analysis adding an additional 29 basis points of cost. Importantly, we find that outside the group of high derivative rollers, the average cost of using technical analysis is small and not statistically significant. Our paper makes three contributions to the behavioral finance literature on individual investors. First, we find that the choices of investors in our data using technical analysis are consistent with the behavior of subjects in experimental studies who use price charts. Second, we find that the behavioral traits of investors using technical analysis are similar to those which the literature links to excessive optimism and overconfidence. Third, we find that high derivative rollers who use technical analysis and have speculation as their primary investment objective exhibit the same behavioral traits as investors who favor lottery stocks. The paper is organized as follows. Section 2 describes the literature we use to develop hypotheses, which we set out in Section 3. Section 4 introduces our data and methods. Section 5 reports results. Section 6 discusses some broader issues. Section 7 concludes.

Since the seminal work of LLS (1974, 1976, 1977, 1978a,b, 1980), which analyzed data from the 1960s, the study of how technical analysis impacts individual investors has received little attention in the finance literature. Yet, the investment landscape has changed dramatically since those times, especially with the advent of online trading through discount brokerage. By analyzing data on Dutch investors during the period 2000–2006 who used an online discount broker, we find that technical analysis impacts the portfolios of individual investors in economically important ways. To the best of our knowledge, ours is the first paper to isolate the impact of using technical analysis from other factors such as speculation as an investment objective, turnover, concentration, and demographic variables such as gender. Relative to the use of other strategies, we find that technical analysis is associated with greater portfolio concentration, more turnover, less betting on trends, more options trading, a higher ratio of nonsystematic risk to total risk, lower gross and net returns, and lower risk-adjusted returns. We estimate that for our data, technical analysis costs investors on average approximately 50 basis points per month in raw returns from poor portfolio selection decisions, and 20 basis points from additional transaction costs. Notably, the impact of technical analysis is concentrated among high derivative rollers, where the costs are much higher: 140 basis points in raw returns, and 29 basis points from additional transaction costs. In terms of both magnitude and statistical significance, the effects of technical analysis are strongly confined to investors who are high derivative rollers. Basically, high derivative rollers who do not use technical analysis earn 56 basis points per month of risk-adjusted return less than investors who are not high derivative rollers, irrespective of whether the latter use technical analysis. Our study shows that additionally technical analysis dramatically accentuates the poor average performance of high derivative rollers by -130 basis points. Notably, high derivative rollers hold riskier portfolios than non-high derivative rollers, and report having larger risk appetites. Our results add to the literature on technical analysis, by providing empirical evidence to support the remarks in Neely (1997) that technical analysis is not suitable for individual investors, even though it can be suitable for professional investors. Our results add to the literature documenting that for individual investors, high turnover and concentration are manifestations of excessive optimism and overconfidence (see Barber and Odean, 2000, Glaser and Weber, 2007, Goetzmann and Kumar, 2008 and Anderson, 2013). We find strong effects associated with high derivatives turnover, suggesting that excessively optimistic, overconfident investors are much more inclined to use technical analysis than other investors. In this regard, we find evidence of causality from high derivative rolling to technical analysis, with high derivative rollers being almost twice as inclined to use technical analysis than other investors. Half of all high derivative rollers use technical analysis. Our results add to the literature documenting that individual investors are prone to invest in lottery-like securities that feature high risk and negative risk-adjusted returns (see Kumar, 2009 and Han and Kumar, 2013). We find that technical analysis is the high octane gasoline that speculative high derivative rollers use to fuel their lottery-like trading. In this regard, the incremental impact of technical analysis on the risk-adjusted returns to high derivative rollers is 468 basis points per month less for speculators than for non-speculators. Our results add to the literature about individual investors and price trends, for which there is both an empirical literature (Grinblatt and Keloharju, 2001) and an experimental literature (Andreassen, 1987 and Andreassen, 1988). Both provide evidence that for trends based on horizons longer than two days, investors are prone to bet on reversals. However, the experimental literature also suggests that when subjects are provided with information that purportedly explains the cause of the trend, they bet on continuation, not reversal. Our results indicate that investors who use technical analysis bet less on reversals than other investors. We find that the reported use of technical analysis is greater in our data than in the older LLS data, 32% vs. 27%, suggesting that costly behaviors have increased over time. The proportion of investors who use technical analysis by itself is more than double in our data than in the LLS data, 9% vs. 4%. Most notable is the marked reduction in the number of investors who report using fundamental analysis, with the percentage in our data being less by a factor of roughly two thirds, 20% vs. 65%. The general advice from behavioral finance for individual investors is that they restrict their attempts to beat the market, and instead invest most of their portfolios as if markets were efficient (Shefrin, 2000). Doing so helps them avoid shooting themselves in the foot as a result of falling prey to their own psychological vulnerabilities. Our findings suggest that this advice is apt for individual investors using technical analysis, especially if they are trading options online through a discount broker.