,

CHAPTER 12

TECHNICAL ANALYSIS

Barry M. Sine, CFA

Miami Beach, FL, U.S.A.

Robert A. Strong, CFA

Orono, ME, U.S.A.

LEARNING OUTCOMES

After completing this chapter, you will be able to do the following:

• Explain the principles of technical analysis, its applications, and its underlying assumptions.
• Discuss the construction and interpretation of different types of technical analysis charts.
• Demonstrate the uses of trend, support and resistance lines, and change in polarity.
• Identify and interpret common chart patterns.
• Discuss common technical analysis indicators: price-based indicators, momentum oscillators, sentiment, and flow of funds.
• Explain the use of cycles by technical analysts.
• Discuss the key tenets of Elliott Wave Theory and the importance of Fibonacci numbers.
• Describe intermarket analysis as it relates to technical analysis and asset allocation.

1. INTRODUCTION

Technical analysis has been used by traders and analysts for centuries, but it has only recently achieved broad acceptance among regulators and the academic community. This chapter gives a brief overview of the field, compares technical analysis with other schools of analysis, and describes some of the main tools in technical analysis. Some applications of technical analysis are subjective. That is, although certain aspects, such as the calculation of indicators, have specific rules, the interpretation of findings is often subjective and based on the long-term context of the security being analyzed. This aspect is similar to fundamental analysis, which has specific rules for calculating ratios, for example, but introduces subjectivity in the evaluation phase.

2. TECHNICAL ANALYSIS: DEFINITION AND SCOPE

Technical analysis is a form of security analysis that uses price and volume data, which is often graphically displayed, in decision making. Technical analysis can be used for securities in any freely traded market around the globe. A freely traded market is one where willing buyers trade with willing sellers without external intervention or impediment. Prices are the result of the interaction of supply and demand in real time. Technical analysis is used on a wide range of financial instruments, including equities, bonds, commodity futures, and currency futures.

The underlying logic of technical analysis is simple:

• Supply and demand determine prices.
• Changes in supply and demand cause changes in prices.
• Prices can be projected with charts and other technical tools.

Technical analysis of any financial instrument does not require detailed knowledge of that instrument. As long as the chart represents the action in a freely traded market, a technician does not even need to know the name or type of the security to conduct the analysis. Technical analysis can also be applied over any time frame—from short-term price movements to long-term movements of annual closing prices. Trends that are apparent in short-term charts may also appear over longer time frames. Because fundamental analysis is more time consuming than technical analysis, investors with short-term time horizons, such as traders, tend to prefer technical analysis—but not always. For example, fundamental analysts with long time frames often perform technical analysis to time the purchase and sale of the securities they have analyzed.

2.1. Principles and Assumptions

Technical analysis can be thought of as the study of collective investor psychology, or sentiment. Prices in any freely traded market are set by human beings or their automated proxies (such as computerized trading programs), and price is set at the equilibrium between supply and demand at any instant in time. Various fundamental theorists have proposed that markets are efficient and rational, but technicians believe that humans are often irrational and emotional and that they tend to behave similarly in similar circumstances.

Although fundamental data are key inputs into the determination of value, these data are analyzed by humans, who may be driven, at least partially, by factors other than rational factors.1 Human behavior is often erratic and driven by emotion in many aspects of one’s life, so technicians conclude that it is unreasonable to believe that investing is the one exception where humans always behave rationally. Technicians believe that market trends and patterns reflect this irrational human behavior. Thus, technical analysis is the study of market trends or patterns. And technicians believe the trends and patterns tend to repeat themselves and are, therefore, somewhat predictable. So, technicians rely on recognition of patterns that have occurred in the past in an attempt to project future patterns of security prices.

Another tenet of technical analysis is that the market reflects the collective knowledge and sentiment of many varied participants and the amount of buying and selling activity in a particular security. In a freely traded market, only those market participants who actually buy or sell a security have an impact on price. And the greater the volume of a participant’s trades, the more impact that market participant will have on price. Those with the best information and most conviction have more say in setting prices than others because the informed traders trade higher volumes. To make use of their information, however, they must trade. Technical analysis relies on knowledgeable market participants putting this knowledge to work by trading in the market, thereby influencing prices and volume. Without trading, the information is not captured in the charts. Arguably, although insider trading is illegal for a variety of reasons, it improves the efficiency of technical analysis.

Trades determine volume and price. The impact occurs instantaneously and frequently anticipates fundamental developments correctly. So, by studying market technical data—price and volume trends—the technician is seeking to understand investor sentiment. The technician is benefiting from the wide range of knowledge of market participants and the collective conclusion of market participants about a security. In contrast, the fundamental analyst must wait for the release of financial statements to conduct financial statement analysis, so a time lag occurs between the market’s activities and the analyst’s conclusions.

Charles Dow, creator in 1896 of what is now known as the Dow Jones Industrial Average, described the collective action of participants in the markets as follows:

The market reflects all the jobber knows about the condition of the textile trade; all the banker knows about the money market; all that the best-informed president knows of his own business, together with his knowledge of all other businesses; it sees the general condition of transportation in a way that the president of no single railroad can ever see; it is better informed on crops than the farmer or even the Department of Agriculture. In fact, the market reduces to a bloodless verdict all knowledge bearing on finance, both domestic and foreign.

A similar notion was expressed by George A. Akerlof and Robert J. Shiller:

To understand how economies work and how we can manage them and prosper, we must pay attention to the thought patterns that animate people’s ideas and feelings, their animal spirits. We will never really understand important economic events unless we confront the fact that their causes are largely mental in nature.2

Market participants use many inputs and analytical tools before trading. Fundamental analysis is a key input in determining security prices, but it is not the only one. Technical analysts believe that emotions play a role. Investors with a favorable fundamental view may nonetheless sell a financial instrument for other reasons, including pessimistic investor sentiment, margin calls, and requirements for their capital—for example, to pay for a child’s college tuition. Technicians do not care why market participants are buying or selling, just that they are doing so.

Some financial instruments have an associated income stream that contributes to the security’s intrinsic value. Bonds have regular coupon payments, and equity shares may have underlying cash flows or dividend streams. A fundamental analyst can adjust these cash flows for risk and use standard time value of money techniques to determine a present value. Other assets, such as a bushel of wheat, gallon of crude oil, or ounce of silver, do not have underlying financial statements or an income stream, so valuation models cannot be used to derive their fundamental intrinsic values. For these assets, technical analysis is the only form of analysis possible. So, whereas fundamental analysis is widely used in the analysis of fixed-income and equity securities, technical analysis is widely used in the analysis of commodities, currencies, and futures.

Market participants attempt to anticipate economic developments and enter into trades to profit from them. Technicians believe that security price movements occur before fundamental developments unfold—certainly before they are reported. This belief is reflected in the fact that stock prices are one of the 12 components of the National Bureau of Economic Research’s Index of Leading Economic Indicators. A key tenet of technical analysis is that the equity market moves roughly six months ahead of inflection points in the broad economy.

2.2. Technical and Fundamental Analysis

Technical analysis and fundamental analysis are both useful and valid, but they approach the market in different ways. Technicians focus solely on analyzing markets and the trading of financial instruments. Fundamental analysis is a much wider field, encompassing financial and economic analysis as well as analysis of societal and political trends. Technicians analyze the result of this extensive fundamental analysis in terms of how it affects market prices. A technician’s analysis is derived solely from price and volume data, whereas a fundamental equity analyst analyzes a company and incorporates data that are external to the market and then uses this analysis to predict security price movements. As the quotation from Dow in Section 2.1 illustrates, technical analysis assumes that all of the factors considered by a fundamental analyst are reflected in the price of a financial instrument through buying and selling activity.

A key distinction between technical analysis and fundamental analysis is that the technician has more concrete data, primarily price and volume data, to work with. The financial statements analyzed by fundamental analysts are not objective data but are the result of numerous estimates and assumptions that have been added together to arrive at the line items in the financial statements. Even the cash line on a balance sheet is subject to corporate management’s opinion about which securities are liquid enough to be considered “cash.” This opinion must be agreed to by auditors and, in many countries, regulators (who sometimes differ with the auditors). Financial statements are subject to restatements because of such issues as changes in accounting assumptions and even fraud. But the price and volume data used in technical analysis are objective. When the data become subject to analysis, however, both types of analysis become subjective because judgment is exercised when a technician analyzes a price chart and when a fundamental analyst analyzes an income statement.

Fundamental analysis can be considered to be the more theoretical approach because it seeks to determine the underlying long-term (or intrinsic) value of a security. Technical analysis can be considered to be the more practical because a technician studies the markets and financial instruments as they exist, even if trading activity appears, at times, to be irrational. Technicians seek to project the level at which a financial instrument will trade, whereas fundamental analysts seek to predict where it should trade.

Being a fundamental analyst can be lonely if the analyst is the first to arrive at a fundamental conclusion, even though it is correct, because deviations from intrinsic value can persist for long periods. The reason these deviations may persist is that it takes buying activity to raise (or lower) the price of a security in a freely traded market.

A drawback of technical analysis is that technicians are limited to studying market movements and do not use other predictive analytical methods, such as interviewing the customers of a subject company, to determine future demand for a company’s products. Technicians study market trends and are mainly concerned with a security’s price trend: Is the security trading up, down, or sideways? Trends are driven by collective investor psychology, however, and can change without warning. Additionally, it can take some time for a trend to become evident. Thus, technicians may make wrong calls and have to change their opinions. Technicians are better at identifying market moves after the moves are already under way.

Moreover, trends and patterns must be in place for some time before they are recognizable, so a key shortcoming of technical analysis is that it can be late in identifying changes in trends or patterns. This shortcoming mirrors a key shortcoming of fundamental analysis in that securities often overshoot fundamental fair values in an uptrend and undershoot fundamental fair values in a downtrend. Strictly relying on price targets obtained by fundamental analysis can lead to closing profitable investment positions too early because investors may irrationally bid securities prices well above or well below intrinsic value.

Fundamental analysis is a younger field than technical analysis because reliable fundamental data are a relatively new phenomenon. In contrast, the first recorded use of technical analysis was in Japan in the 1700s, where it was used to analyze trading in the rice market. The Japanese developed a detailed field of technical analysis with their own chart design and patterns. These tools were translated and widely understood outside Japan only in the 1980s.

Western use of technical analysis was pioneered by Dow, who was also the first editor of the Wall Street Journal, in the 1890s. At the time, publicly traded companies were under no requirement to release their financial information even to shareholders, and insider trading was common and legal. Dow created the Dow Jones Industrial Average and the Dow Jones Railroad Average (now the Transportation Average) as a proxy to gauge the health of the economy, because fundamental data were not available. By his logic, if industrial stocks were doing well, industrial companies themselves must be doing well and if railroad stocks were doing well, railroad companies must be doing well. And if both manufacturers and the companies that transported goods to market were prospering, the economy as a whole must be prospering.

Not until the Securities Exchange Act of 1934 were public companies in the United States required to regularly file financial statements that were available to the public. In that year, Benjamin Graham published his seminal work, Security Analysis, and three years later, he and several others founded one of the first societies devoted to fundamental analysis, the New York Society of Security Analysts.3 Fundamental analysis quickly overtook technical analysis in terms of acceptance by practitioners, regulators, and academics.

Acceptance of technical analysis by practitioners was revived in the 1970s with the creation of the Market Technicians Association in New York and the International Federation of Technical Analysts a few years later. Only in the last decade, however, has the field started to achieve widespread acceptance by regulators and academics. An important impediment to acceptance by academics is the difficulty of capturing the subjectivity involved in technical analysis. The human brain can recognize, analyze, and interpret technical information that is difficult for statistical computer models to recognize and test.

Although technical analysis can be applied to any freely traded security, it does have its limits. In markets that are subject to large outside manipulation, the application of technical analysis is limited. For example, the central banks of many countries intervene in their currency markets from time to time to maintain exchange rate stability. Interestingly, traders claim to have been able to successfully predict interventions in some countries, especially those where the central bank is itself using technical analysis. Technical analysis is also limited in illiquid markets, where even modestly sized trades can have an inordinate impact on prices. For example, in considering a thinly traded American Depositary Receipt (ADR), analyzing the more heavily traded local security frequently yields a better analysis.4 Another example of when technical analysis may give an incorrect reading is in the case of a company that has declared bankruptcy and announced that its shares will have zero value in a restructuring. A positive technical trend may appear in such cases as investors who hold short positions buy shares to close out their positions.

A good example of when technical analysis is a superior tool to fundamental analysis is in the case of securities fraud, such as occurred at Enron Corporation and WorldCom. These companies were issuing fraudulent financial statements, but many fundamental analysts continued to hold favorable views of the companies’ equity securities even as the share prices declined. Simultaneously, a small group of investors came to the opposite view and expressed this view through high-volume sales of the securities. The result was clearly negative chart patterns that could then be discerned by technical analysis.

3. TECHNICAL ANALYSIS TOOLS

The primary tools used in technical analysis are charts and indicators. Charts are the graphical display of price and volume data, and the display may be done in a number of ways. Charts are then subjected to various analyses, including the identification of trends, patterns, and cycles. Technical indicators include a variety of measures of relative price level—for example, price momentum, market sentiment, and funds flow. We will discuss charts first.

3.1. Charts

Charts are an essential component of the technical analyst’s toolkit. Charts provide information about past price behavior and provide a basis for inferring likely future price behavior. A variety of charts can be useful in studying the markets. The selection of the chart to use in technical analysis is determined by the intended purpose of the analysis.

3.1.1. Line Chart

Line charts are familiar to all types of analysts and are a simple graphic display of price trends over time. Usually, the chart is a plot of data points, such as share price, with a line connecting these points. Line charts are typically drawn with closing prices as the data points. The vertical axis (y-axis) reflects price level, and the horizontal axis (x-axis) is time. Even though the line chart is the simplest chart, an analyst can quickly glean information from this chart.

The chart in Exhibit 12-1 is a quarterly chart of the FTSE 100 Index from 1984 through mid-2009. Up years and down years are clearly evident. The strong rally from 1984 through 1999 and the market decline from late 1999 to late 2002 are also clearly visible. The 2003–2007 rally did not exceed the high reached in 1999, which suggests that investors were not willing to pay as high a price for stocks on the London Stock Exchange during that rally as they were in the prior rally. This information provides a broad overview of investor sentiment and can lead to further analysis. Importantly, the analyst can access and analyze this information quickly. Collecting and analyzing the full array of data normally incorporated in fundamental analysis would take much longer.

3.1.2. Bar Chart

A line chart has one data point per time interval. A bar chart, in contrast, has four bits of data in each entry—the high and low price encountered during the time interval plus the opening and closing prices. Such charts can be constructed for any time period, but they are customarily constructed from daily data.

As Exhibit 12-2 shows, a vertical line connects the high and low price of the day; a cross-hatch to the right indicates the closing price, and a cross-hatch to the left indicates the opening price. The appeal of this chart is that the analyst immediately gets a sense of the nature of that day’s trading. A short bar indicates little price movement during the day; that is, the high, low, and close were near the opening price. A long bar indicates a wide divergence between the high and the low for the day.

Exhibit 12-3 shows daily performance of the Brazilian Bovespa Index (BVSP) from late 2007 through late 2009. The top part provides the price open, close, high, and low; the bottom part shows volume, which will be discussed in Section 3.1.6. The downturn in the second half of 2008 is obvious, but also notable are the extreme price movements in the fourth quarter of 2008. There were 40 trading days from 29 September to 24 November. On 20 of those days, the closing value of the index changed from the previous close by at least 4 percent, a huge move by historical standards. During the same period, the average daily price range (high to low) was 7 percent, compared with 3.7 percent in the previous two months. This potentially important information would not be captured in a line chart.

3.1.3. Candlestick Chart

Candlestick charts trace their roots to Japan, where technical analysis has been in use for centuries. Like a bar chart, a candlestick chart also provides four prices per data point entry: the opening and closing prices and the high and low prices during the period. As shown in Exhibit 12-4, a vertical line represents the range through which the security price traveled during the time period. The line is known as the wick or shadow. The body of the candle is shaded if the opening price was higher than the closing price, and the body is clear if the opening price was lower than the closing price.

Exhibit 12-5 shows a weekly candlestick chart for Companhia Vale do Rio Doce for the period 1 January through 15 June 2009.

The advantage of the candlestick chart over the bar chart is that price moves are much more visible in the candlestick chart, which allows faster analysis. The bar chart indicates market volatility only by the height of each bar, but in candlestick charts, the difference between opening and closing prices and their relationship to the highs and lows of the day are clearly apparent. Compare the sixth candle with the twelfth in Exhibit 12-5. In the sixth candle, the analyst can see significant volatility because the high of the day and low of the day are so far apart. The stock opened near the low of the day and closed near the high, suggesting a steady rally during the day. In contrast, the twelfth candle shows no difference between the high and low, and the shares opened and closed at the same price, creating a cross pattern. In Japanese terminology used in candlestick charting, this pattern is called a doji. The doji signifies that after a full day of trading, the positive price influence of buyers and the negative price influence of sellers exactly counteracted each other, which tells the analyst that this market is in balance. If a doji occurs at the end of a long uptrend or downtrend, it signals that the trend will probably reverse.

3.1.4. Point and Figure Chart

Point and figure charts were widely used in the United States in the early 1900s and were favored because they were easy to create and update manually in the era before computers. As with any technical analysis tool, these charts can be used with equities, fixed-income securities, commodities, or foreign exchange.

Where the point and figure chart originated is unclear; the chart is referred to in a number of books in the United States dating back to 1898. The methodology evolved until 1934 when the first book was published on the topic: The Point and Figure Method of Anticipating Stock Price Movements by Victor de Villiers and Owen Taylor. With the advent of powerful charting software and Internet web sites, complex chart types, such as the candlestick chart, have become more popular. But point and figure charts still offer tremendous value if one knows their limitations and their advantages. The key reason this knowledge is necessary, as explained next, is that point and figure charts are constructed differently from other charts; they have a clear focus on entry and exit levels but no focus on holding periods.

As illustrated in Exhibit 12-6, a point and figure chart is drawn on a grid and consists of columns of X’s alternating with columns of O’s. Neither time nor volume is represented on this type of chart, and the horizontal axis represents the number of changes in price, not time. Movement along the horizontal axis does reflect the passage of time, but not in any uniform fashion. The analyst makes entries on a point and figure chart only when the price changes by the “box size,” which is explained below. This lack of a normal time dimension is perhaps the most unusual characteristic of a point and figure chart.

To construct a point and figure chart, the analyst must determine both the box size and the reversal size. Box size refers to the change in price represented by the height of each box (boxes are generally square, but the width has no meaning). In Exhibit 12-6, the box size is HK$1. The reversal size is used to determine when to create a new column. In Exhibit 12-6, the reversal size is three, meaning a reversal in price of three or more boxes.

Although a point and figure chart can be constructed in several ways, these charts are always drawn on graph paper to facilitate seeing the “columns and rows” nature of the data. The vertical axis measures discrete increments of price. For example, an analyst in Europe might draw a €1 chart, a €2 chart, or any other increment. In a €1 chart, boxes would be €1 apart (e.g., €40, €41, €42), whereas in a €2 chart they would be €2 apart (€40, €42, €44). The most commonly used box size is 1 unit of currency, which is used when prices range from 20 to 100 per share of the currency.

The next decision the technician needs to make is the reversal size. The most common size is three, meaning a reversal in price of three or more boxes (€3 in the case of a box size of €1). This use of a multibox reversal helps eliminate “noise” in the price data. (Noise refers to short-term trading volatility that does not alter the long-term trend of the security.)

In a point and figure chart, X represents an increase in price and O represents a decline in price. In constructing a chart, the technician draws an X in a column of boxes every time the security price closes up by the amount of the box size. (Ideally, all security prices are considered on an intraday basis, but this practice has given way to using closing prices only.) If the price increases by twice the box size, the technician draws two X’s to fill in two boxes, one on top of the other. The technician fills in more boxes for larger price moves. The resulting column starts at the opening price level and extends to the closing price level. As long as the security keeps closing at higher prices, the technician keeps filling in boxes with X’s, which makes the column higher and higher. If the price does not increase by at least the box size, no indication is made on the chart. Thus, in some cases, the chart is not updated for long periods, but no indication of this passage of time is made on the chart.

The reversal size determines when to create a new column. In the case of a €1 box size, and three-box reversal size, a decline of €3 or more would result in the technician shifting to the next column over and beginning a column of O’s. The first box to be filled in is to the right and below the highest X in the prior column. The technician then fills in an O to bring the column down to the price level at the close. Again, each filled-in box (if the box size is €1) represents a €1 decline in the security price. As long as the downtrend continues, without a €3 increase in price, the technician continues adding O’s to the column below the prior O’s. A reversal in the downtrend by at least the amount of the reversal size prompts the technician to move to the next column and begin drawing a series of X’s again. Computer technology makes the process easy, but many technicians prefer to keep point and figure charts on their wall and update them manually because doing so provides a vivid reminder of the market trend.

Point and figure charts are particularly useful for making trading decisions because they clearly illustrate price levels that may signal the end of a decline or advance. They also clearly show price levels at which a security may frequently trade. In using the point size and reversal size to make trading decisions, for uptrends, or columns of X’s, the practitioner would maintain long positions. The reversal size could be considered the amount of loss that would prompt the closing of a long position and the establishment of a new short position. The larger the reversal size, the fewer columns in the chart and the longer uptrends and downtrends will run.

The box size can be varied in relation to the security price. For a security with a very low price—say, below €5—a €1 box size might mean few or no updates on the chart because the price would only rarely change by this amount. Thus, the technician could reduce the box size to cents. For highly priced securities, much larger box sizes could be used. The reversal size is a multiple of the box size, so if the box size is changed, the reversal size changes. Practitioners who want fewer columns or trade signals can use a large reversal size.

Analysis of a point and figure chart is relatively straightforward as long as the technician understands its construction and limitations. The chart is relatively simple, and repeated high and low prices are evident. Congestion areas, where a security trades up and down in a narrow range, are evidenced by a series of short columns of X’s and O’s spanning roughly the same price range. Major, sustained price moves are represented by long columns of X’s (when prices are moving up) or O’s (when prices are moving down).

3.1.5. Scale

For any chart—line, bar, or candlestick—the vertical axis can be constructed with either a linear scale (also known as an arithmetic scale) or a logarithmic scale, depending on how you want to view the data. With a logarithmic scale, equal vertical distances on the chart correspond to an equal percentage change. A logarithmic scale is appropriate when the data move through a range of values representing several orders of magnitude (e.g., from 10 to 10,000); a linear scale is better suited for narrower ranges (e.g., prices from $35 to $50). The share price history of a particular company, for instance, is usually best suited to a linear scale because the data range is usually narrow.

The horizontal axis shows the passage of time. The appropriate time interval depends on the nature of the underlying data and the specific use of the chart. An active trader, for instance, may find 10-minute, 5-minute, or even tick-by-tick data useful, but other technical analysts may prefer daily or weekly data. In general, the greater the volatility of the data, the greater the likelihood that an analyst can find useful information in more-frequent data sampling.

Consider Exhibits 12-7 and 12-8, which both show the yearly history of the Dow Jones Industrial Average (DJIA) from 1928 to 2010. Plotting the index on a linear scale, as in Exhibit 12-7, makes it difficult to gather much information from the first 50 years of the data series. Analysts can see a slight uptrend but not much else. The eye is drawn to the bull market of the 1980s, the subsequent dot-com bubble, and the recent era of the subprime crisis. When plotted on a logarithmic scale, as in Exhibit 12-8, however, many people would find that the data tell a more comprehensive story. The Great Depression of the 1930s stands out, but over the following 75 years, the data follow a relatively stable upward trend.

3.1.6. Volume

Volume is an important characteristic that is included at the bottom of many charts; see, for example, Exhibit 12-3. Volume is used to assess the strength or conviction of buyers and sellers in determining a security’s price. For example, on a daily price chart, below the price section would be a column chart showing the volume traded for that day.

Some technicians consider volume information to be crucial. If volume increases during a time frame in which price is also increasing, that combination is considered positive and the two indicators are said to “confirm” each other. The signal would be interpreted to mean that over time, more and more investors are buying the financial instrument and they are doing so at higher and higher prices. This pattern is considered a positive technical development.

Conversely, if volume and price diverge—for example, if a stock’s price rises while its volume declines—the implication is that fewer and fewer market participants are willing to buy that stock at the new price. If this trend in volume continues, the price rally will soon end because demand for the security at higher prices will cease. Exhibit 12-9 shows a bar chart for Toronto-Dominion Bank (TD Bank) with volume displayed separately.

3.1.7. Time Intervals

Most of the chart examples in this chapter are daily price charts in that they show the price and volume on a daily basis. Daily frequency is not required, however, because charts can be constructed by using any time interval. For short-term trading, the analyst can create charts with one-minute or shorter intervals. For long-term investing, the analyst can use weekly, monthly, or even annual intervals. The same analytical approach applies irrespective of the time interval. Using long intervals allows the analyst to chart longer time periods than does using short time intervals for the simple reason that long intervals contain fewer data points, so a longer time frame can be presented on the chart. Using short intervals allows the analyst to see more detail. A useful step for many analysts is to begin the analysis of a security with the chart for a long time frame, such as a weekly or monthly chart, and then construct charts with shorter and shorter time intervals, such as daily or hourly charts.

3.1.8. Relative Strength Analysis

Relative strength analysis is widely used to compare the performance of a particular asset, such as a common stock, with that of some benchmark—such as, in the case of common stocks, the FTSE 100, the Nikkei 225, or the S&P 500 Index—or the performance of another security. The intent is to show out- or underperformance of the individual issue relative to some other index or asset. Typically, the analyst prepares a line chart of the ratio of two prices, with the asset under analysis as the numerator and with the benchmark or other security as the denominator. A rising line shows the asset is performing better than the index or other stock; a declining line shows the opposite. A flat line shows neutral performance.

Suppose a private investor is researching two investment ideas she read about. Harley-Davidson Motor Company (HOG) is a well-known motorcycle company; Rodman and Renshaw (RODM) is a small investment bank. The investor wants to determine which of these two has been the stronger performer (relative to the S&P 500) over the past few months. Exhibit 12-10 shows relative strength lines for the two stocks for the first six months of 2009. Each point on the relative strength plot is simply the ratio of a share price to the S&P 500. For example, on 9 March 2009, HOG closed at US$8.42 and the S&P 500 closed at $676.53. The relative strength data point is, therefore, 8.42/676.53, or 0.0124. On 27 April, HOG closed at US$19.45, with the S&P 500 at $857.51. The relative strength value is 19.45/857.51, or 0.0227, nearly double the 9 March value.

The units on the vertical axis are not significant; the ratio is a function of the relative prices of the assets under consideration. The important information is how the ratio has changed. This type of chart allows an analyst to make a visual determination of that change. As Exhibit 12-10 illustrates, Harley-Davidson was a strong performer in March and April but lagged the index beginning in May. In contrast, the stock of Rodman and Renshaw began a significant rise in mid-May that outperformed the market average.

3.2. Trend

The concept of a trend is perhaps the most important aspect of technical analysis. Trend analysis is based on the observation that market participants tend to act in herds and that trends tend to stay in place for some time. A security can be considered to be in an upward trend, a downward trend, a sideways trend, or no apparent trend. Not all securities are in a trend, and little useful forecasting information can be gleaned from technical analysis when a security is not in a trend. Not every chart will have obvious or clear implications, so the analyst must avoid the temptation to force a conclusion from every chart and thus reach a wrong interpretation.

An uptrend for a security is when the price goes to higher highs and higher lows. As the security moves up in price, each subsequent new high is higher than the prior high and each time there is a retracement, which is a reversal in the movement of the security’s price, it must stop at a higher low than the prior lows in the trend period. To draw an uptrend line, a technician draws a line connecting the lows of the price chart. Major breakdowns in price, however, when the price drops through and below the trendline by a significant amount (many technicians use 5–10 percent below the trendline) indicate that the uptrend is over and may signal a further decline in the price. Minor breakthroughs below previous lows simply call for the line to be moderately adjusted over time. Time is also a consideration in trends: The longer the security price stays below the trendline, the more meaningful the breakdown is considered to be.

In an uptrend, the forces of demand are greater than the forces of supply. So, traders are willing to pay higher and higher prices for the same asset over time. Presumably, the strong demand indicates that investors believe the intrinsic value of the security is increasing.

A downtrend is when a security makes lower lows and lower highs. As the security moves down in price, each subsequent new high must be lower than the prior high and each time there is a retracement, it must stop at a lower low than the prior lows in the trend period. To draw a downtrend line, a technician draws a line connecting the highs of the price chart. Major breakouts above the downtrend line (e.g., 5–10 percent) indicate that the downtrend is over and a rise in the security’s price may occur. And as with an uptrend, the longer the security price stays above the trendline, the more meaningful the breakout is considered to be.

In a downtrend, supply is overwhelming demand. Over time, sellers are willing to accept lower and lower prices to exit long positions or enter new short positions. Both motives of the sellers generally indicate deteriorating investor sentiment about the asset. However, selling may be prompted by factors not related to the fundamental or intrinsic value of the stock. For example, investors may be forced to sell to meet margin calls in their portfolios. From a purely technical standpoint, the reason is irrelevant. The downtrend is assumed to continue until contrary technical evidence appears. Combining fundamental analysis with technical analysis in such a case, however, might reveal a security that has attractive fundamentals but a currently negative technical position. In uptrends, however, a security with an attractive technical position but unattractive fundamentals is rare because most buying activity is driven by traders who expect the security price to increase in the future. The rare exception is covering short positions after a sizable decline in the share price.

A security may trade in a fairly narrow range, moving sideways on the price chart without much upward or downward movement. This pattern indicates a relative balance between supply and demand. A technical analyst may not expect to profit from long or short trades in such securities but might devise profitable option strategies for short-term investors with the ability to accept the risks.

Exhibit 12-11 shows the application of trend analysis. Depicted is an uptrend line for the shares of China Mobile Limited. Note that through late 2007, every rally took the shares to a new high whereas sell-offs stopped at increasingly higher levels. The first sign of trouble came in the spring of 2008 when the rally terminated at a lower price point than the prior rally of late 2007. This movement was followed by the shares breaking through the trendline.

The chart in Exhibit 12-11 covers roughly seven years and would most likely be used by investors with a long time horizon. Investors with a shorter horizon might use a chart with a shorter time frame and would thus obtain a different trendline as well as a different trendline breakdown.

Two concepts related to trend are support and resistance. Support is defined as a low price range in which buying activity is sufficient to stop the decline in price. It is the opposite of resistance, which is a price range in which selling is sufficient to stop the rise in price. The psychology behind the concepts of support and resistance is that investors have come to a collective consensus about the price of a security. Support and resistance levels can be sloped lines, as in trendlines, or horizontal lines.

A key tenet of support and resistance as a part of technical analysis is the change in polarity principle, which states that once a support level is breached, it becomes a resistance level. The same holds true for resistance levels; once breached, they become support levels. For example, if the price of a security never rises above SFr10 over a long period of time and begins to decline each time it reaches this level but then finally breaks through this level by a significant amount, the point to which the price rises becomes a support level.

Support and resistance levels are commonly round numbers. Support indicates that at some price level, investors consider a security to be an attractive investment and are willing to buy, even in the wake of a sharp decline (and for resistance, at some level, investors are not willing to buy, even in an uptrend). The fact that these price points tend to be round numbers strongly suggests that human sentiment is at work.

One of the most widely publicized examples of support and resistance is when the DJIA broke through the 10,000 mark in 1999, shown in Exhibit 12-12. Previously, 10,000 had been viewed as a resistance line, but from 1999 through the end of the chart in 2001, 10,000 served as a support level.

3.3. Chart Patterns

Chart patterns are formations that appear in price charts that create some type of recognizable shape. Common patterns appear repeatedly and often lead to similar subsequent price movements. Thus, the identification and analysis of chart patterns is a common aspect of technical analysis used to predict security prices. An important connection to understand is that patterns form as a result of the behavior of market participants and that these patterns represent graphical depictions of the collective psychology of the market at a given time.

The recurring patterns that appear in charts can be used as a basis for market forecasting. The reason chart patterns have predictive value is that they are graphic representations of human trading activity and human behavior is frequently repeated, especially trading activity that is driven by fear (in market sell-offs) or hope and greed (as evidenced in bubbles—that is, rallies that extend well beyond valuation levels that would be derived by fundamental values). An example of a rally driven by greed is the recent real estate bubble, which took home prices to unsustainably high levels. This bubble started a few years after the Internet stock bubble of the 1990s, which also took prices to unsustainably high levels. In bubbles, investors, driven by hope and greed, drive the price of an asset to irrationally high levels, in the expectation that another buyer will be willing to pay an even higher price for the asset. The housing bubble was notable because it so closely followed the Internet stock bubble, despite all that had been written about the “irrational exuberance” of the Internet bubble of the 1990s.

Chart patterns can be divided into two categories: reversal patterns and continuation patterns. These terms refer to the trend for the security in question prior to the formation of the pattern. The most important concept to understand in using chart patterns is that without a clear trend in place prior to the pattern, the pattern has no predictive value. This aspect is frequently forgotten by investors who are so eager to identify and use patterns that they forget the proper application of charts.

3.3.1. Reversal Patterns

As the name implies, a reversal pattern signals the end of a trend, a change in direction of the financial instrument’s price. Evidence that the trend is about to change direction is obviously important, so reversal patterns are noteworthy.

3.3.1.1. Head and Shoulders

Perhaps the most widely recognized reversal pattern is the head and shoulders pattern. The pattern consists of three segments. Volume is an important characteristic in interpreting this pattern. Because head and shoulders indicates a trend reversal, a clear trend must exist prior to the formation of the pattern in order for the pattern to have predictive validity. For a head and shoulders pattern, the prior trend must be an uptrend. Later, we will discuss the inverse head and shoulders pattern (preceded by a downtrend).

Exhibit 12-13 depicts a head and shoulders pattern for Marvell Technology Group during 2006. The three parts of the pattern are as follows:

• Left shoulder: This part appears to show a strong rally, with the slope of the rally being greater than the prior uptrend, on strong volume. The rally then reverses back to the price level where it started, forming an inverted V pattern, but on lower volume.
• Head: The head is a more pronounced version of the left shoulder. A rally following the first shoulder takes the security to a higher high than the left shoulder by a significant enough margin to be clearly evident on the price chart. Volume is typically lower in this rally, however, than in the one that formed the first, upward side of the left shoulder. This second rally also fails, with price falling back to the same level at which the left shoulder began and ended. This price level is called the neckline. This price level also will be below the uptrend line formed by connecting the low prices in the uptrend preceding the beginning of the head and shoulders pattern. This head pattern is the first signal that the rally may be coming to an end and that a reversal may be starting.
• Right shoulder: The right shoulder is a mirror image (or close to a mirror image) of the left shoulder but on lower volume, signifying less buying enthusiasm. The price rallies up to roughly the same level as the first shoulder, but the rally reverses at a lower high price than the rally that formed the head.

Rarely will an analyst see a perfectly formed head and shoulders pattern; variations include two tops on the shoulders or on the head. The head, however, should rise to a higher price level than either shoulder, whereas the shoulders should be roughly symmetrical. In terms of the neckline price level, the first rally should begin at this level and the left shoulder and head should also decline to roughly this level. But necklines may not always form exactly horizontal lines. These imperfect variations make this (and other) technical patterns difficult for quantitative analysts or academicians to model, but the human brain can detect the pattern even if it is imperfectly formed.

Volume is important in analyzing head and shoulders patterns. A new high in price at the top of the head without a new high in volume signals fewer bullish market participants. When one indicator is making a new high (or low) but another is not, this situation is called divergence. In divergence, the right shoulder will have even lower volume, signaling that buying interest or demand is tapering off and will soon be overwhelmed by supply. The result will be a price decline.

Once the head and shoulders pattern has formed, the expectation is that the share price will decline down through the neckline price. Technicians tend to use filtering rules to make sure that a clear breakdown of the neckline has occurred. These rules may take the form of waiting to trade until the price falls to a meaningful level below the neckline (3 percent or 5 percent are commonly used) and/or a time limit for the price to remain below the neckline before trading; when a daily price chart is used, the rule may be several days to a week. Prices commonly rebound to the neckline levels, even after a decline has exceeded the filter levels. Prices generally stop, however, at or around the neckline. The neckline was a support level, and under the change in polarity principle, once a support level is breached, it becomes a resistance level.

3.3.1.2. Inverse Head and Shoulders

The head and shoulders pattern can also form upside down and act as a reversal pattern for a preceding downtrend. The three parts of the inverse head and shoulders are as follows:

• Left shoulder: This shoulder appears to show a strong decline, with the slope of the decline greater than the prior downtrend, on strong volume. The rally then reverses back to the price level where it started, forming a V pattern, but on lower volume.
• Head: The head is a more pronounced version of the left shoulder. Another decline follows but on diminishing volume, which takes the price to a lower low than the prior shoulder by a significant enough margin that it is clearly evident on the price chart. This second decline also reverses, with price rising to the same level at which the left shoulder began and ended. This price level, the neckline, will also be above the uptrend line formed by connecting the high prices in the downtrend preceding the beginning of the inverse head and shoulders pattern. This pattern is the first signal that the decline may be coming to an end and that a reversal may be near.
• Right shoulder: The right shoulder is roughly a mirror image of the left shoulder but on lower volume, signifying less selling enthusiasm. The price declines down to roughly the same level as the first shoulder, but the rally reverses at a higher low price than the rally that formed the head.

3.3.1.3. Setting Price Targets with Head and Shoulders Pattern

As with all technical patterns, the head and shoulders pattern must be analyzed from the perspective of the security’s long-term price trend. The rally that happened before the formation of the pattern must be large enough for there to be something to reverse. The stronger and more pronounced the rally was, the stronger and more pronounced the reversal is likely to be. Similarly, once the neckline is breached, the security is expected to decline by the same amount as the change in price from the neckline to the top of the head. If the preceding rally started at a price higher than the neckline, however, the correction is unlikely to bring the price lower than the price level at the start of the rally. Because a head and shoulders formation is a bearish indicator (i.e., a technician would expect the previously established uptrend to end and a downtrend to commence), a technician would seek to profit by shorting the security under analysis. When attempting to profit from the head and shoulders pattern, a technician will often use the price differences between the head and the neckline to set a price target, which is the price at which the technician anticipates closing the investment position. The price target for the head and shoulders pattern is calculated as follows:

For example, in Exhibit 12-14, the high price reached at the top of the head is roughly $37 and the neckline formed at roughly $27 for a difference of $10. So a technician would expect the price to decline to a level $10 below the neckline, or to $17; that is,

3.3.1.4. Setting Price Targets with Inverse Head and Shoulders Pattern

Calculating price targets for inverse head and shoulders patterns is similar to the process for head and shoulders patterns, but in this case, because the pattern predicts the end of a downtrend, the technician calculates how high the price is expected to rise once it breaches the neckline. Exhibit 12-15 illustrates an inverse head and shoulders pattern.

For an inverse head and shoulders pattern, the formula is similar to a head and shoulders pattern:

For example, in the price chart in Exhibit 12-15, the low price reached at the bottom of the head is roughly US$7,197 and the neckline formed at roughly US$9,050. The target can thus be found as $9,050 + (9,050 – $7,197) = $10,903. In this case, a technician might have taken a long position in the summer of 2003 with the hope of eventually exiting the position at about US$10,903 for a profit.

3.3.1.5. Double Tops and Bottoms

A double top is when an uptrend reverses twice at roughly the same high price level. Typically, volume is lower on the second high than on the first high, signaling a diminishing of demand. The longer the time is between the two tops and the deeper the sell-off is after the first top, the more significant the pattern is considered to be. Price targets can be calculated from this pattern in a manner similar to the calculation for the head and shoulders pattern. For a double top, price is expected to decline below the low of the valley between the two tops by at least the distance from the valley low to the high of the double tops.

Double bottoms are formed when the price reaches a low, rebounds, and then sells off back to the first low level. Exhibit 12-16 depicts a double bottom pattern for Time Warner. Technicians use the double bottom to predict a change from a downtrend to an uptrend in security prices. For double bottoms, the price is expected to appreciate above the peak between the two bottoms by at least the distance from the valley lows to the peak.

The reason these patterns are significant is that they show that at some price point, investors step in to reverse trends that are under way. For an uptrend, a double top implies that at some price point, enough traders are willing to either sell positions (or enter new short positions) that their activities overwhelm and reverse the uptrend created by demand for the shares. A reasonable conclusion is that this price level has been fundamentally derived and that it represents the intrinsic value of the security that is the consensus of investors. With double bottoms, if a security ceases to decline at the same price point on two separate occasions, the analyst can conclude that the market consensus is that at that price point, the security is now cheap enough that it is an attractive investment.

3.3.1.6. Triple Tops and Bottoms

Triple tops consist of three peaks at roughly the same price level, and triple bottoms consist of three troughs at roughly the same price level. A triple top for Rockwell Automation during 1999 is shown in Exhibit 12-17.

One of the challenges in double-top and triple-top patterns, and one of the valid criticisms of technical analysis in general, is that an analyst cannot know which pattern will result until after the fact. For example, after the broad equity market sell-off in the first quarter of 2009, a number of investment professionals were quoted as calling for a “retest of the lows”—in technical terms, a double bottom.

There is no evidence that market corrections (or rallies) must end with a double bottom (or double top in the case of an uptrend), and there is no generally accepted technical theory that predicts whether a low will be repeated once or even twice before a reversal occurs. A double bottom is considered to be a more significant pattern than a single bottom because traders have stepped in on two occasions to halt declines. However, traders have no way to determine whether a double top or bottom will be followed by a third top or bottom. Triple tops and triple bottoms are rare, but when they occur, they are more significant reversal patterns than double tops or double bottoms. On three separate occasions, traders stepped in to sell or buy shares with enough volume to end a rally or decline under way at the time. Nevertheless, the greater the number of times the price reverses at the same level, and the greater the time interval over which this pattern occurs, the greater the significance of the pattern.

3.3.2. Continuation Patterns

A continuation pattern is used to predict the resumption of a market trend that was in place prior to the formation of a pattern. From a supply-and-demand standpoint, a continuation pattern indicates a change in ownership from one group of investors to another. For example, if a positive trend was in place prior to a pattern and then one group of investors begins selling, the negative impact on price is quickly offset by other investors buying, so the forces of supply and demand go back and forth in terms of their impact on price. But neither has an overwhelming advantage. This type of pattern is often called “a healthy correction” because the long-term market trend does not change and because while one set of investors is seeking to exit, they are replaced by another set of investors willing to take their positions at roughly the same share price.

3.3.2.1. Triangles

Triangle patterns are a type of continuation pattern. They come in three forms, symmetrical triangles, ascending triangles, and descending triangles. A triangle pattern forms as the range between high and low prices narrows, visually forming a triangle. In old terminology, triangles were referred to as “coils” (which was also synonymous with “springs”) because a triangle was considered analogous to a spring being wound up tighter and tighter and storing energy that would at some point be released. In a triangle, a trendline connects the highs and a trendline connects the lows. As the distance between the highs and lows narrows, the trendlines meet, forming a triangle. In a daily price chart, a triangle pattern usually forms over a period of several weeks.

In an ascending triangle, as shown in Exhibit 12-18, the trendline connecting the high prices is horizontal and the trendline connecting the low prices forms an uptrend. What this pattern means is that market participants are selling the stock at the same price level over a period of time, putting a halt to rallies at the same price point, but that buyers are getting more and more bullish and stepping in at increasingly higher prices to halt sell-offs instead of waiting for further price declines. An ascending triangle typically forms in an uptrend. The horizontal line represents sellers taking profits at around the same price point, presumably because they believe that this price represents the fundamental, intrinsic value of the security. The fact that the rally continues beyond the triangle may be a bullish signal; it means that another set of investors is presumably willing to buy at an even higher price because their analysis suggests the intrinsic value of the security is higher. Alternatively, the fundamental facts themselves may have changed; that is, the security’s fundamental value may be increasing over time. The technician does not care which explanation is true; the technician is relying solely on the information conveyed by the security price itself, not the underlying reason.

In the descending triangle, shown in Exhibit 12-19, the low prices form a horizontal trendline and the high prices form a series of lower and lower highs. Typically, a descending triangle will form in a downtrend. At some point in the sell-offs, buyers appear with enough demand to halt sell-offs each time they occur, at around the same price. Again, this phenomenon may be the result of fundamental analysts believing that the security has reached a price where it represents a significant discount to its intrinsic value and these analysts step in and buy. As the triangle forms, each rally ceases at a lower and lower high price point, suggesting that the selling demand is exerting greater price influence than the buying demand.

In a symmetrical triangle, the trendline formed by the highs angles down and the trendline formed by the lows angles up, both at roughly the same angle, forming a symmetrical pattern. Exhibit 12-20 contains a symmetrical triangle formed by the price for Transocean in early 2000. What this triangle indicates is that buyers are becoming more bullish while, simultaneously, sellers are becoming more bearish, so they are moving toward a point of consensus. Because the sellers are often dominated by long investors exiting positions (as opposed to short sellers creating new short positions), the pressure to sell diminishes once the sellers have sold the security. Thus, the pattern ends in the same direction as the trend that preceded it, either uptrend or downtrend.

The term “measuring implication” refers to the height of a triangle, as illustrated with a dark vertical bar in Exhibit 12-20. The measuring implication is derived by calculating the difference in price from the two trendlines at the start of the triangle. Once the pattern is broken and the price breaks through one of the trendlines that form the triangle, the analyst expects the price to move by at least the amount of the breakthrough above or below the trendline. Typically, price breaks out of a triangle pattern between halfway and three-quarters of the way through the pattern. The longer the triangle pattern persists, the more volatile and sustained the subsequent price movement is likely to be.

3.3.2.2. Rectangle Pattern

A rectangle pattern is a continuation pattern formed by two parallel trendlines, one formed by connecting the high prices during the pattern, and the other formed by the lows. Exhibit 12-21 shows two rectangle patterns. As is the case with other patterns, the rectangle pattern is a graphical representation of what has been occurring in terms of collective market sentiment. The horizontal resistance line that forms the top of the rectangle shows that investors are repeatedly selling shares at a specific price level, bringing rallies to an end. The horizontal support line forming the bottom of the rectangle indicates that traders are repeatedly making large enough purchases at the same price level to reverse declines. The support level in a bullish rectangle is natural because the long-term trend in the market is bullish. The resistance line may simply represent investors taking profits. Conversely, in a bearish rectangle, the support level may represent investors buying the security. Again, the technician is not concerned with why a pattern has formed, only with the likely next price movement once the price breaks out of the pattern.

3.3.2.3. Flags and Pennants

Flags and pennants are considered minor continuation patterns because they form over short periods of time—on a daily price chart, typically over a week. They are similar to each other and have the same uses. A flag is formed by parallel trendlines, in the same way that most countries’ flags are rectangular and create a parallelogram. Typically, the trendlines slope in a direction opposite to the trend up to that time; for example, in an uptrend, they slope down. A pennant formation is similar except that the trendlines converge to form a triangle, similar to the pennants of many sports teams or pennants flown on ships. The key difference between a triangle and pennant is that a pennant is a short-term formation whereas a triangle is a long-term formation.

The expectation for both flags and pennants is that the trend will continue after the pattern in the same direction it was going prior to the pattern. The price is expected to change by at least the same amount as the price change from the start of the trend to the formation of the flag or pennant. In Exhibit 12-22, a downtrend begins at point A, which is $104. At point B, which is $70, a pennant begins to form. The distance from point A to point B is $34. The pennant ends at point C, which is $76. The price target is $76 minus $34, which is $42, the line labeled D.

3.4. Technical Indicators

The technical analyst uses a variety of technical indicators to supplement the information gleaned from charts. A technical indicator is any measure based on price, market sentiment, or funds flow that can be used to predict changes in price. These indicators often have a supply-and-demand underpinning; that is, they measure how potential changes in supply and demand might affect a security’s price.

3.4.1. Price-Based Indicators

Price-based indicators somehow incorporate information contained in the current and past history of market prices. Indicators of this type range from simple (e.g., a moving average) to complex (e.g., a stochastic oscillator).

3.4.1.1. Moving Average

A moving average is the average of the closing price of a security over a specified number of periods. Moving averages smooth out short-term price fluctuations, giving the technician a clearer image of market trend. Technicians commonly use a simple moving average, which weights each price equally in the calculation of the average price. Some technicians prefer to use an exponential moving average (also called an exponentially smoothed moving average), which gives the greatest weight to recent prices while giving exponentially less weight to older prices.

The number of data points included in the moving average depends on the intended use of the moving average. A 20-day moving average is commonly used because a month contains roughly 20 trading days. Also, 60 days is commonly used because it represents a quarter year (three months) of trading activity.

Moving averages can be used in conjunction with a price trend or in conjunction with one another. Moving averages are also used to determine support and resistance.

Because a moving average is less volatile than price, this tool can be used in several ways. First, whether price is above or below its moving average is important. A security that has been trending down in price will trade below its moving average, and a security that has been trending up will trade above its moving average. Second, the distance between the moving-average line and price is also significant. Once price begins to move back up toward its moving-average line, this line can serve as a resistance level. The 65-day moving-average line is commonly cited in the press, and when the price approaches the moving-average line, many investors become concerned that a rally will stall, so they sell the security.

Two or more moving averages can be used in conjunction. Exhibit 12-23 shows the price chart of Gazprom SP European Depositary Receipts (EDRs) on the Frankfurt Stock Exchange overlaid with 20-day and 60-day EDR moving averages for late 2007 to mid-2009.5 Note that the longer the time frame used in the creation of a moving average, the smoother and less volatile the line. Investors often use moving-average crossovers as a buy or sell signal. When a short-term moving average crosses from underneath a longer-term average, this movement is considered bullish and is termed a golden cross. Conversely, when a short-term moving average crosses from above a longer-term moving average, this movement is considered bearish and is called a dead cross. In the case shown in Exhibit 12-23, a trading strategy of buying on golden crosses and selling on dead crosses would have been profitable.

Moving averages are easy to construct, and simple trading rules can be derived for using them. Computers can optimize what time lengths to set when using two moving averages. This optimization may take the form of changing the number of days included in each moving average or adding filter rules, such as waiting several days after a trade signal is given to make a trade. Reasons for optimization include the desire to manage capital drawdowns, to maximize gains, or to minimize losses. Once the moving average is optimized, even if a profitable trading system is devised for that security, the strategy is unlikely to work for other securities, especially if they are dissimilar. Also, as market conditions change, a previously optimized trading system may no longer work.

3.4.1.2. Bollinger Bands

Market veteran John Bollinger combined his knowledge of technical analysis with his knowledge of statistics to create an indicator called Bollinger Bands. Bollinger Bands consist of a moving average plus a higher line representing the moving average plus a set number of standard deviations from average price (for the same number of periods as used to calculate the moving average) and a lower line that is a moving average minus the same number of standard deviations. Exhibit 12-24 depicts Bollinger Bands for the Gazprom EDR.

The more volatile the security being analyzed becomes, the wider the range becomes between the two outer lines or bands. Similar to moving averages, Bollinger Bands can be used to create trading strategies that can be easily computerized and tested. A common use is as a contrarian strategy, in which the investor sells when a security price reaches the upper band and buys when it reaches the lower band. This strategy assumes that the security price will stay within the bands.

This type of strategy is likely to lead to a large number of trades, but it also limits risk because the trader can quickly exit unprofitable trades. In the event of a sharp price move and a change in trend, however, a contrarian strategy based on Bollinger Bands would be unprofitable. So, long-term investors might actually buy on a significant breakout above the upper boundary band because a major breakout would imply a change in trend likely to persist for some time. The long-term investor would sell on a significant breakout below the lower band. In this strategy, significance would be defined as breaking above or below the band by a certain percentage (say, 5 percent or 10 percent) and/or for a certain period of time (say, a week for a daily price chart). Again, such rules can easily be computerized and tested.

3.4.2. Momentum Oscillators

One of the key challenges in using indicators overlaid on a price chart is the difficulty of discerning changes in market sentiment that are out of the ordinary. Momentum oscillators are intended to alleviate this problem. They are constructed from price data, but they are calculated so that they either oscillate between a high and low (typically 0 and 100) or oscillate around a number (such as 0 or 100). Because of this construction, extreme highs or lows are easily discernible. These extremes can be viewed as graphic representations of market sentiment when selling or buying activity is more aggressive than historically typical. Because they are price based, momentum oscillators also can be analyzed by using the same tools technicians use to analyze price, such as the concepts of trend, support, and resistance.

Technicians also look for convergence or divergence between oscillators and price. Convergence is when the oscillator moves in the same manner as the security being analyzed, and divergence is when the oscillator moves differently from the security. For example, when price reaches a new high, this sign is considered bullish, but if the momentum oscillator being used does not also reach a new high at the same time, this pattern is divergence. It is considered to be an early warning of weakness, an indication that the uptrend may soon end.

Momentum oscillators should be used in conjunction with an understanding of the existing market (price) trend. Oscillators alert a trader to overbought or oversold conditions. In an overbought condition, market sentiment is unsustainably bullish. In an oversold condition, market sentiment is unsustainably bearish. In other words, the oscillator range must be considered separately for every security. Some securities may experience wide variations, and others may experience only minor variations.

Oscillators have three main uses. First, oscillators can be used to determine the strength of a trend. Extreme overbought levels are warning signals for uptrends, and extreme oversold levels are warning signals for downtrends. Second, when oscillators reach historically high or low levels, they may be signaling a pending trend reversal. For oscillators that move above and below 0, crossing the 0 level signals a change in the direction of the trend. For oscillators that move above and below 100, crossing the 100 level signals a change in the direction of the trend. Third, in a non-trending market, oscillators can be used for short-term trading decisions—that is, to sell at overbought levels and to buy at oversold levels.

3.4.2.1. Momentum or Rate of Change Oscillator

The terms momentum oscillator and rate of change oscillator are synonymous. “Rate of change” is often abbreviated ROC. The ROC oscillator is calculated by taking the most recent closing price, subtracting the closing price from a prior date that is a set number of days in the past, and multiplying the result by 100:

where

When the ROC oscillator crosses zero in the same direction as the direction of the trend, this movement is considered a buy or sell signal. For example, if the ROC oscillator crosses into positive territory during an uptrend, it is a buy signal. If it enters into negative territory during a downtrend, it is considered a sell signal. The technician will ignore crossovers in opposition to the trend because the technician must always first take into account the general trend when using oscillators.

An alternative method of constructing this oscillator it to set it so that it oscillates above and below 100, instead of 0, as follows:

This approach is shown in Exhibit 12-25 for Toyota Motor Corporation.

In Exhibit 12-25, the calculation method for the ROC oscillator for Toyota stock, traded on the Tokyo Stock Exchange, is for the oscillator to move around 100 and x is 12 days. Note that for this stock, the ROC oscillator tends to maintain a range between ¥85 and ¥115. So episodes when the oscillator moves outside this range are of particular interest to the technician. An extreme high means that the stock has posted its highest gain in any 12-day period at this point, and an extreme low reading means it has posted its greatest loss over any 12-day period. When investors bid up the price of a security too rapidly, the indication is that sentiment may be unduly bullish and the market may be overbought. Exhibit 12-25 shows that overbought levels of the ROC oscillator coincide with temporary highs in the stock price. So, those levels would have been signals to sell the stock. The other notable aspect of Exhibit 12-25 is the divergence when the share price hit a new low in December 2008 but the ROC oscillator did not. This divergence would have been a bullish signal and would have been interpreted to mean that, although the share price hit a new low, investor sentiment was actually higher than it had been previously. In itself, this information would not have been enough to warrant buying the shares because a downtrend in price was still in place, but it alerted the technician to the fact that the trend might end soon. The technician could then look for further indication of the trend’s end and, with confirmation, might buy the stock.

3.4.2.2. Relative Strength Index

A relative strength index (RSI) is computed over a rolling time period.6 It graphically compares a security’s gains with its losses over the set period. The creator of the RSI, Welles Wilder, suggested a 14-day time period, and this period is generally the period used in most technical analysis software. The technician should understand that this variable can be changed and that the optimal time range should be determined by how the technician intends to use the RSI information. Factors that influence selection of the time period are similar to those that influence the selection of a time period for moving averages. Short time periods (such as 14 days) provide information about short-term price behavior. If 200 days is used, this short-term information will be smoothed out and, perhaps, will not be apparent at all.

RSI is a momentum oscillator and is not to be confused with the charting method called “relative strength analysis,” in which the ratio of two security prices is plotted over time. The RSI provides information on whether or not an asset is overbought. The formula for the RSI is not intuitive and is best understood with an example. The formula is:

Exhibit 12-26 shows closing prices for Ford Motor Company during the month of June 2009.

During this time, markets were still rebounding from the subprime crisis; automobile company stocks were unusually volatile and, to some speculators, presented interesting short-term trading opportunities. Suppose a trader decided to compute an RSI for the month of June. It would be a 22-day RSI with 21 price changes—11 up, 9 down, and 1 unchanged. To calculate the RSI, the trader would sum the 11 up changes, which sum to US$1.45. The down changes total –US$1.51; the absolute value drops the minus sign. The ratio of these two numbers is 0.96, so the RSI is

The index construction forces the RSI to lie within 0 and 100. A value above 70 represents an overbought situation. Values below 30 suggest the asset is oversold. Again, as is the case with most technical tools, an analyst cannot simply learn the default settings and use them in every case. The 30–70 range is a good rule of thumb, but because the oscillator is a measure of volatility, less volatile stocks (such as utilities) may normally trade in a much narrower range. More volatile stocks (such as small-capitalization technology stocks) may trade in a wider range. The range also does not have to be symmetrical around 50. For example, in an uptrend, one might see a range of 40–80 but in downtrends, a range of 20–60.

The RSI measure often appears at the bottom or top of a price chart. Exhibit 12-27 shows a candlestick chart of Ford stock in 2009 with the corresponding RSI.

The candlestick chart of Ford stock prices in Exhibit 12-27 illustrates several aspects of the use of an RSI. For example, because the RSI oscillator was higher than 70 on 23 March so the stock was overbought at that time, a simple reading of the chart might have led to the conclusion that the trader should sell the stock. Doing so, however, would have caused the trader to miss a significant advance in the shares. A more careful technical analysis that took into account the trend would have indicated that the stock was in an uptrend, so RSI readings above 70 could be expected.

Because RSI is a price-based oscillator, the trader can also apply trend lines to analyze it. Note in Exhibit 12-27 that both the share price and the RSI oscillator were in uptrends from February until April but that the RSI uptrend was broken on 15 April, a potential warning that the uptrend in price might also break downward. In June, the share price broke its uptrend support line.

3.4.2.3. Stochastic Oscillator

The stochastic oscillator is based on the observation that in uptrends, prices tend to close at or near the high end of their recent range and in downtrends, they tend to close near the low end. The logic behind these patterns is that if the shares of a stock are constantly being bid up during the day but then lose value by the close, continuation of the rally is doubtful. If sellers have enough supply to overwhelm buyers, the rally is suspect. If a stock rallies during the day and is able to hold on to some or most of those gains by the close, that sign is bullish.

The stochastic oscillator oscillates between 0 and 100 and has a default setting of a 14-day period, which, again, might be adjusted for the situation as we discussed for the RSI. The oscillator is composed of two lines, called %K and %D, that are calculated as follows:

where

and

Analysts should think about the %D in the same way they would a long-term moving-average line in conjunction with a short-term line. That is, %D, because it is the average of three %K values, is the slower moving, smoother line and is called the signal line. And %K is the faster moving line. The %K value means that the latest closing price (C) was in the %K percentile of the high–low range (L14 to H14).

The default oversold–overbought range for the stochastic oscillator is based on reading the signal line relative to readings of 20 and 80, but warnings about always using the default range for the RSI oscillator also apply in the case of the stochastic oscillator. In fact, noted technician Constance Brown has coined a term called the “stochastics default club” to refer to neophyte technicians who trade based solely on these defaults.7 She has reported being able to develop successful trading strategies by using a time frame shorter than the 14-day default to calculate the stochastic oscillator. Apparently, enough traders are basing trades on the defaults to move the market for certain stocks. So, using shorter time frames than the default, she could trade ahead of the traders in the default stochastic club and generate a profit. Of course, other traders might be tempted to use an even shorter time frame, but there is a drawback to using a short time frame; namely, the shorter the time frame is, the more volatile the oscillator becomes and the more false signals it generates.

The stochastic oscillator should be used with other technical tools, such as trend analysis or pattern analysis. If both methods suggest the same conclusion, the trader has convergence (or confirmation), but if they give conflicting signals, the trader has divergence, which is a warning signal suggesting that further analysis is necessary.

The absolute level of the two lines should be considered in light of their normal range. Movements above this range indicate to a technician an overbought security and are considered bearish; movements below this range indicate an oversold security and are considered bullish. Crossovers of the two lines can also give trading signals the same way crossovers of two moving averages give signals. When the %K moves from below the %D line to above it, this move is considered a bullish short-term trading signal; conversely, when %K moves from above the %D line to below it, this pattern is considered bearish. In practice, a trader can use technical analysis software to adjust trading rules and optimize the calculation of the stochastic oscillator for a particular security and investment purpose (e.g., short-term trading or long-term investing).

The reason technicians use historical data to test their trading rules and find the optimal parameters for each security is that each security is different. The group of market participants actively trading differs from security to security. Just as each person has a different personality, so do groups of people. In effect, the groups of active market participants trading each security are imparting their personality on the trading activity for that security. As this group changes over time, the ideal parameters for a particular security may change.

Exhibit 12-28 provides a good example of how the stochastic oscillator can be used together with trend analysis. The exhibit provides the weekly price chart and stochastic oscillator for Petroleo Brasileiro ADRs, which are traded on the New York Stock Exchange, for June 2008 through June 2009. Note that during the downtrend on the left side of the chart the stochastic oscillator often moved below 20. Each time it reached 80, however, it provided a valid sell signal. When the downtrend ended in November 2008 and an uptrend began, the stochastic oscillator was regularly moving above 80 but each time the %D line moved above %K, a valid buy signal was given.

3.4.2.4. Moving-Average Convergence/Divergence Oscillator

The moving-average convergence/divergence oscillator is commonly referred to as MACD, which is pronounced Mack Dee. The MACD is the difference between a short-term and a long-term moving average of the security’s price. The MACD is constructed by calculating two lines, the MACD line and the signal line:

• MACD line: difference between two exponentially smoothed moving averages, generally 12 and 26 days
• Signal line: exponentially smoothed average of MACD line, generally 9 days

The indicator oscillates around zero and has no upper or lower limit. Rather than using a set overbought–oversold range for MACD, the analyst compares the current level with the historical performance of the oscillator for a particular security to determine when a security is out of its normal sentiment range.

MACD is used in technical analysis in three ways. The first is to note crossovers of the MACD line and the signal line, as discussed for moving averages and the stochastic oscillator. Crossovers of the two lines may indicate a change in trend. The second is to look for times when the MACD is outside its normal range for a given security. The third is to use trend lines on the MACD itself. When the MACD is trending in the same direction as price, this pattern is convergence, and when the two are trending in opposite directions, the pattern is divergence.

Exhibit 12-29 shows a daily price chart of Exxon Mobil (at the top) with the MACD oscillator for March through October of 2005. Note the convergence in the bottoming of both the oscillator and price in May, which provided confirmation of a change in trend. This change was further confirmed by the MACD line crossing above the signal line. A bearish signal was given in September with the change in trend of both price and the oscillator and the crossover of the signal line by the MACD line. The fact that the MACD oscillator was moving up to a level that was unusually high for this stock would have been an early warning signal in September.

3.4.3. Sentiment Indicators

Sentiment indicators attempt to gauge investor activity for signs of increasing bullishness or bearishness. Sentiment indicators come in two forms—investor polls and calculated statistical indices.

3.4.3.1. Opinion Polls

A wide range of services conduct periodic polls of either individual investors or investment professionals to gauge their sentiment about the equity market. The most common of the polls are the Investors Intelligence Advisors Sentiment reports, Market Vane Bullish Consensus, Consensus Bullish Sentiment Index, and Daily Sentiment Index, all of which poll investment professionals, and reports of the American Association of Individual Investors (AAII), which polls individual investors. All but the AAII survey are subscription-based services. Barron’s magazine publishes data from four of these surveys on a weekly basis.

By regularly polling, compiling these data over time, and presenting it graphically, these services provide technicians with an analyzable snapshot of investor sentiment over time. Technicians look at prior market activity and compare it with highs or lows in sentiment, as well as inflection points in sentiment, as a gauge when they are forecasting the future direction of the market.

The most widely used investor polls are all U.S.-based. One reason is that interpretation of the surveys is determined by comparing the survey results with market performance over time. To gauge a survey’s usefulness in predicting major market turns, the survey must have been published over several cycles, and each of the surveys mentioned here, based on U.S. data, has been available for several decades.

3.4.3.2. Calculated Statistical Indices

The other category of sentiment indicators are indicators that are calculated from market data, such as security prices. The two most commonly used are derived from the options market; they are the put/call ratio and the volatility index. Additionally, many analysts look at margin debt and short interest.

The put/call ratio is the volume of put options traded divided by the volume of call options traded for a particular financial instrument. Investors who buy put options on a security are presumably bearish, and investors who buy call options are presumably bullish. The volume in call options is greater than the volume traded in put options over time, so the put/call ratio is normally below 1.0. The ratio is considered to be a contrarian indicator, meaning that higher values are considered bearish and lower values are considered bullish. But, its usefulness as a contrarian indicator is limited except at extreme low or high levels in relation to the historical trading level of the put/call ratio for a particular financial instrument. The actual value of the put/call ratio, and its normal range, differs for each security or market, so no standard definitions of overbought or oversold levels exist. At extreme lows where call option volume is significantly greater than put option volume, market sentiment is said to be so overly positive that a correction is likely. At extreme highs in the put/call ratio, market sentiment is said to be so extremely negative that an increase in price is likely.

The CBOE Volatility Index (VIX) is a measure of near-term market volatility calculated by the Chicago Board Options Exchange. Since 2003, it has been calculated from option prices on the stocks in the S&P 500. The VIX rises when market participants become fearful of an impending market decline. These participants then bid up the price of puts, and the result is an increase in the VIX level. Technicians use the VIX in conjunction with trend, pattern, or oscillator tools, and it is interpreted from a contrarian perspective. When other indicators suggest that the market is oversold and the VIX is at an extreme high, this combination is considered bullish. Exhibit 12-30 shows the VIX from March 2005 to December 2009.

Margin debt is also often used as an indication of sentiment. As a group, investors have a history of buying near market tops and selling at the bottom. When the market is rising and indices reach new highs, investors are motivated to buy more equities in the hope of participating in the market rally. A margin account permits an investor to borrow part of the investment cost from the brokerage firm. This debt magnifies the gains or losses resulting from the investment.

Investor psychology plays an important role in the intuition behind margin debt as an indicator. When stock margin debt is increasing, investors are aggressively buying and stock prices will move higher because of increased demand. Eventually, the margin traders use all of their available credit, so their buying power (and, therefore, demand) decreases, which fuels a price decline. Falling prices may trigger margin calls and forced selling, thereby driving prices even lower.

Brokerage firms must report activity in their customers’ margin accounts, so keeping track of borrowing behavior is relatively easy. Exhibit 12-31 provides a 10-year comparison of margin debt with the S&P 500. The correlation is striking: Rising margin debt is generally associated with a rising index level, and falling margin debt is associated with a falling index level. In fact, for the 113 months shown in Exhibit 12-31, the correlation coefficient between the levels of margin debt and the S&P 500 is 80.2 percent. When margin debt peaked in the summer of 2007, the market also topped out. Margin debt dropped sharply during the latter part of 2008 as the subprime crisis took the market down. Investors began to use borrowed funds again in the first half of 2009 when heavily discounted shares became increasingly attractive. Margin debt was still well below the average of the last decade, but the upturn would be viewed as a bullish sign by advocates of this indicator.

Short interest is another commonly used sentiment indicator. Investors sell shares short when they believe the share prices will decline. Brokerage firms must report short-sale activity, and these statistics are aggregated and reported by the exchanges and the financial press on a monthly basis. The number of shares of a particular security that are currently sold short is called “short interest.” The short interest ratio represents the number of days trading activity represented by short interest. To facilitate comparisons of large and small companies, common practice is to “normalize” this value by dividing short interest by average daily trading volume to get the short interest ratio:

There are differences of opinion about how to interpret short interest as an indicator. It is considered to show market sentiment and to be a contrarian indicator. Some people believe that if a large number of shares are sold short and the short interest ratio is high, the market should expect a falling price for the shares because of so much negative sentiment about them. A counterargument is that, although the short sellers are bearish on the security, the effect of their short sales has already been felt in the security price. The short sellers’ next action will be to buy shares back to cover their short positions. When the short sellers cover their positions, those actions will provide a boost to the share price. Therefore, the short interest ratio constitutes future (and known) demand for the shares.

Regardless of the analyst’s perspective, in Example 12-3, the TD Banknorth short interest ratio of approximately 17 is more noteworthy than the much lower figure for Goldman Sachs.

3.4.4. Flow-of-Funds Indicators

Technicians look at fund flows as a way to gauge the potential supply and demand for equities. Demand can come in the form of margin borrowing against current holdings or cash holdings by mutual funds and other groups that are normally large holders of equities, such as insurance companies and pension funds. The more cash these groups hold, the more bullish is the indication for equities. One caveat in looking at potential sources of demand is that, although these data indicate the potential buying power of various large investor groups, the data say nothing about the likelihood that the groups will buy.

On the supply side, technicians look at new or secondary issuance of stock because these activities put more securities into the market and increase supply.

3.4.4.1. Arms Index

A common flow of funds indicator is the Arms Index, also called the TRIN (for “short-term trading index”).8 This indicator is applied to a broad market (such as the S&P 500) to measure the relative extent to which money is moving into or out of rising and declining stocks. The index is a ratio of two ratios:

When this index is near 1.0, the market is in balance; that is, as much money is moving into rising stocks as into declining stocks. A value above 1.0 means that there is more volume in declining stocks; a value below 1.0 means that most trading activity is in rising stocks. Exhibit 12-32 shows the Arms Index for the S&P 500 on a daily basis for the first six months of 2009. The majority of the points lie above the 1.0 level, suggesting that the market continued to be in a selling mood. Note that the up spikes are associated with large price decreases in the index level and the down spikes reflect the opposite. The trendline shows a slightly negative slope, providing some slight encouragement for the bulls.

3.4.4.2. Margin Debt

The previous section discussed the use of margin debt as an indicator of market sentiment. Margin debt is also widely used as a flow-of-funds indicator because margin loans may increase the purchases of stocks and declining margin balances may force the selling of stocks.

3.4.4.3. Mutual Fund Cash Position

Mutual funds hold a substantial proportion of all investable assets. Some analysts use the percentage of mutual fund assets held in cash as a predictor of market direction. It is called the “mutual fund cash position indicator.” Mutual funds must hold some of their assets in cash in order to pay bills and send redemption checks to account holders. Cash arrives on a daily basis from customer deposits, interest earned, and dividends received. Cash also increases after a fund manager sells a position and holds the funds before reinvesting them. During a bull market, the manager wants to buy shares as quickly as possible to avoid having a cash “drag” hurt the fund’s performance. If prices are trending lower, however, the manager may hold funds in cash to improve the fund’s performance.

Exhibit 12-33 shows year-end mutual fund cash in the United States as a percentage of assets from 1984 through 2008. Over this period, the average cash percentage was 6.8 percent. An analyst’s initial intuition might be that when cash is relatively low, fund managers are bullish and anticipate rising prices but when fund managers are bearish, they conserve cash to wait for lower prices. Advocates of this technical indicator argue exactly the opposite: When the mutual fund cash position is low, fund managers have already bought, and the effects of their purchases are already reflected in security prices. When the cash position is high, however, that money represents buying power that will move prices higher when the money is used to add positions to the portfolio. The mutual fund cash position is another example of a contrarian indicator.

Some analysts modify the value of the cash percentage to account for differences in the level of interest rates. Cash is not sitting in a desk drawer; it is on deposit somewhere earning interest. When interest rates are low, holding cash can be a substantial drag on the fund’s performance if the broad market advances. When interest rates are high, holding cash is less costly.

3.4.4.4. New Equity Issuance

When a company’s owners decide to take a company public and offer shares for sale, the owners want to put those shares on the market at a time when investors are eager to buy. That is, the owners want to offer the shares when they can sell them at a premium price. Premium prices occur near market tops. The new equity issuance indicator suggests that as the number of initial public offerings (IPOs) increases, the upward price trend may be about to turn down.

A supply-and-demand effect is also at work. Putting more shares on the market increases the aggregate supply of shares available for investors to purchase. The investment community has a finite quantity of cash to spend, so an increase in IPOs may be viewed as a bearish factor.

3.4.4.5. Secondary Offerings

Technicians also monitor secondary offerings to gauge potential changes in the supply of equities. Although secondary offerings do not increase the supply of shares, because existing shares are sold by insiders to the general public, they do increase the supply available for trading or the float. So, from a market perspective, secondary offerings of shares have the potential to change the supply-and-demand equation as much as IPOs do.

3.5. Cycles

Over the centuries, technicians have noted recurring cycles of various frequencies in the capital markets. The study of cycles in the markets is part of broader cycle studies that exist in numerous fields of study. Many observed cycles, such as one in U.S. equities tied to the cycle of U.S. presidential elections, have an obvious and rational justification. Other cycles do not. However, why cycles in fields seemingly unrelated to finance, such as astronomy or weather patterns, may influence the economy (and thus the capital markets) may have a logical explanation. For example, sunspots affect weather patterns on earth, which in turn affect agriculture and, therefore, capital markets because they are related to agriculture.

3.5.1. Kondratieff Wave

The longest of the widely recognized cycles was identified by Nikolai Kondratieff in the 1920s. Kondratieff was an economist in the Soviet Union who suggested that Western economies had a 54-year cycle. He traced cycles from the 1780s to the time he published this theory in the 1920s, and the economic depression of the 1930s was consistent with the cycle he identified. His theory was mainly tied to economic cycles and commodity prices, but cycles can also be seen in the prices of equities during the time of his work.

Kondratieff was executed in a Soviet purge in 1938, but his ideas have come into widespread acceptance, particularly since his works were translated into English in the 1980s. Two economists at the London School of Economics, E. H. Phelps Brown and Sheila Hopkins, identified a 50- to 52-year economic cycle in the United Kingdom. Together with Kondratieff, credit should be given to two Dutch economists, Jacob van Gelderen and Samuel de Wolff, who wrote about a 50- to 60-year economic cycle but published their work earlier, in 1913. Their work came to light only recently, however, so the long 54-year economic cycle is known as the Kondratieff Wave or K Wave.

3.5.2. 18-Year Cycle

The 18-year cycle is interesting because three 18-year cycles make up the longer 54-year Kondratieff Wave. The 18-year cycle is most often mentioned in connection with real estate prices, but it can also be found in equities and other markets.

3.5.3. Decennial Pattern

The decennial pattern is the pattern of average stock market returns (based on the DJIA) broken down on the basis of the last digit in the year. Years ending with a 0 have had the worst performance, and years ending with a 5 have been by far the best. The DJIA was up every year ending in a 5 from 1885 until 1995, but it declined 0.6 percent in 2005.

3.5.4. Presidential Cycle

This cycle in the United States connects the performance of the DJIA with presidential elections. In this theory, years are grouped into categories on the basis of whether they were election years or the first, second, or third year following an election. The third year is the year prior to the next election. The third year shows the best performance; in fact, the DJIA experienced a positive return in every pre-election year from 1943 through 2007. One explanation for this outcome is that with so many politicians up for re-election, they inject stimulus into the economy in an attempt to improve their chances to be re-elected.9 Election years are also usually positive years for the stock market, but with less consistency. Postelection years and the so-called midterm year have the worst performance.

These long cycles are important to keep in mind when using other technical analysis tools. However, the long cycles described here and other theories about long cycles present a number of problems. The primary problem is the small sample size. Only 56 presidential elections have been held in the United States, and only 4 completed Kondratieff cycles have occurred in U.S. history. Another problem is that even with the small number of cycles, the data do not always fit the cycle theory, and when they do, that fit may not be obvious.

4. ELLIOTT WAVE THEORY

In a theory proposed by R. N. Elliott in 1938, the market moves in regular, repeated waves or cycles. He identified and categorized these waves and wrote in detail about aspects of market cycles. Elliott was an accountant by training, but in 1929, after he contracted a progressive intestinal illness at age 58 while working in Latin America, he was forced to retire. Then, he turned his attention to a detailed study of equity prices in the United States.

A decade later, in 1938, he published his findings in a book titled The Wave Principle. In developing the concept that the market moves in waves, Elliott relied heavily on Charles Dow’s early work. Elliott described how the market moved in a pattern of five waves moving up in a bull market in the following pattern: 1 = up, 2 = down, 3 = up, 4 = down, and 5 = up. He called this wave the “impulse wave.” The impulse wave was followed by a corrective wave with three components: a = down, b = up, and c = down.

When the market is a bear market, as defined in Dow Theory—that is, with both of Dow’s major indices in bear markets—the downward movements are impulse waves and are broken into five waves with upward corrections broken into three subwaves.

Elliott also noted that each wave could be broken down into smaller and smaller subwaves.

The longest of the waves is called the “grand supercycle” and takes place over centuries. Elliott traced grand supercycles back to the founding of the United States, and his successors have continued his work. Each grand supercycle can be broken down into subcycles until ending with the “subminuette,” which unfolds over several minutes. The major cycles are:

• Grand supercycle
• Supercycle
• Cycle
• Primary
• Intermediate
• Minor
• Minute
• Minuette
• Subminuette

An important aspect of Elliott’s work is that he discovered that market waves follow patterns that are ratios of the numbers in the Fibonacci sequence. Leonardo Fibonacci was an eleventh-century Italian mathematician who explained this sequence in his book Liber Abaci, but the sequence was known to mathematicians as far back as 200 B.C.E. in India. The Fibonacci sequence starts with the numbers 0, 1, 1, and then each subsequent number in the sequence is the sum of the two preceding numbers:

Elliott was more interested in the ratios of the numbers in the sequence because he found that the ratio of the size of subsequent waves was generally a Fibonacci ratio. The ratios of one Fibonacci number to the next that Elliott considered most important are the following:

He also noticed that the ratio of a Fibonacci sequence number to its preceding number is important:

These ratios converge around 1.618. In mathematics, 1.618 is called the “golden ratio,” and it can be found throughout nature—in astronomy, biology, botany, and many other fields. It is also widely used in art and architecture. The ancient Egyptians built the pyramids on the basis of this ratio, and the ancient Greeks used it widely.

As noted, Elliott numbered the impulse waves 1–5 and the corrective waves, a, b, and c. Exhibit 12-34 depicts the impulse and corrective waves in a bull market.

Elliott described the characteristics of each wave. Note the following, as shown in Exhibit 12-34:

• Wave 1 starts as a basing pattern and displays an increase in price, volume, and breadth.10 Wave 1 consists of five smaller waves.
• Wave 2 moves down, retracing much of the gain in Wave 1 but not all of it. Common percentage retracements are Fibonacci ratios, such as 50 percent or 62 percent. Wave 2 never erases all of the gains from Wave 1. Wave 2 consists of three smaller waves.
• Wave 3 moves above the high of the first wave and has strong breadth, volume, and price movement. Most of the price movement in an uptrend typically occurs in Wave 3. Wave 3 consists of five smaller waves. Wave 3 often moves prices 1.68 times higher than the length of Wave 1, which is a Fibonacci ratio.
• Wave 4 is, again, a correction, and the ratio of the change in price during this wave to the price change during the third wave is also generally a Fibonacci ratio. Wave 4 commonly reverses 38 percent of the gain in Wave 3. Wave 5 is also an up wave. Generally, the price movement in Wave 5 is not as great as that in Wave 3. The exception to the rule is that Wave 5 may become extended, as when euphoria overtakes the market. Wave 5 consists of five smaller waves.

After Wave 5 is completed, the market traces out a series of three corrective waves, labeled a, b, and c in Exhibit 12-34.

• Wave a is a down wave in a bull market; Wave a itself breaks down into three waves.
• Wave b is an upward movement and breaks down into five waves. Wave b is a false rally and is often called a “bull trap.”
• Wave c is the final corrective wave. In a bull market, it does not move below the start of the prior Wave 1 pattern. Wave c breaks down into three subwaves.

This description of the waves applies to bull markets; in bear markets, the impulse waves are labeled A through E and the corrective waves are labeled 1, 2, and 3. Waves in the direction of the trend consist of five subwaves, and counterwaves consist of three subwaves.

In practice, a good deal of time is required to become proficient with Elliott Wave Theory. Wave counts may not become evident at first, and Elliotticians often have to renumber their wave counts on the basis of changes in market trends. This theory is widely used, however, and the patterns Elliott described can still be observed today.

As a technician begins to make initial judgments on wave counts, the next step is to draw lines representing Fibonacci ratios on the charts. These lines alert the technician to the levels at which trends may change in the future and can be used in conjunction with other technical tools for forecasting. Positive price movements generally take prices up by some Fibonacci ratio of prior highs (e.g., 1.5 or 1.62), and price declines generally reverse prices by a Fibonacci ratio (e.g., 0.50 or 0.667). Elliott Wave Theory is used in practice with Dow Theory, trend analysis, pattern analysis, and oscillator analysis to provide a sense of the general trend in the market. As Elliott’s nine cycles imply, Elliott Wave Theory can be applied in both very short term trading as well as in very long term economic analysis, as is the case with most tools used in technical analysis.

5. INTERMARKET ANALYSIS

Intermarket analysis is a field within technical analysis that combines analysis of major categories of securities—namely, equities, bonds, currencies, and commodities—to identify market trends and possible inflections in a trend. Intermarket analysis also looks at industry subsectors, such as the nine sectors the S&P 500 is divided into, and the relationships among the major stock markets of countries with the largest economies, such as the New York, London, and Tokyo stock exchanges.

Intermarket analysis relies heavily on the field of economic analysis for its theoretical underpinning. The field was pioneered by John Murphy with his 1991 book Intermarket Technical Analysis. Murphy noted that all markets are interrelated and that these relationships are strengthening with the globalization of the world economy.11

Stock prices are affected by bond prices. High bond prices are a positive for stock prices since this means low interest rates. Lower interest rates benefit companies with lower borrowing costs and lead to higher equity valuations in the calculation of intrinsic value using discounted cash flow analysis in fundamental analysis. Thus rising bond prices are a positive for stock prices, and declining bond prices are a bearish indicator.

Bond prices impact commodity prices. Bond prices move inversely to interest rates. Interest rates move in proportion to expectations to future prices of commodities or inflation. So declining bond prices are a signal of possible rising commodity prices.

Currencies impact commodity prices. Most commodity trading is denominated in US dollars and so prices are commonly quoted in US dollars. As a result, a strong dollar results in lower commodity prices and vice versa.

In intermarket analysis, technicians often look for inflection points in one market as a warning sign to start looking for a change in trend in a related market. To identify these intermarket relationships, a commonly used tool is relative strength analysis, which charts the price of one security divided by the price of another.

Exhibit 12-35 shows the relative price of 10-year U.S. Treasury bonds compared with the S&P 500. The rise in T-bond price relative to the S&P 500 can be clearly seen. The inflection point in this chart occurs in March 2009. This point would signal that the time had come to move investments from bonds to stocks.

Exhibit 12-36 is a relative strength chart depicting the ratio between the S&P 500 and commodity prices. It shows a clear top and reversal of trend in December 2008. This inflection point shows U.S. stocks weakening relative to commodities and would indicate that allocating funds away from the U.S. stocks and into commodities might be appropriate.

In addition to the preceding comparisons, once an asset category has been identified, relative strength analysis can be used to identify the strongest performing securities in a sector. For example, if commodities look promising, an investor can analyze each of the major commodities relative to a broad commodity index in order to find the strongest commodity.

Intermarket analysis can also be used to identify sectors of the equity market to invest in—often in connection with technical observations of the business cycle at any time. The equities of certain industry sectors tend to perform best at the beginning of an economic cycle. These sectors include utilities, financials, consumer nondurables, and transportation stocks. As an economic recovery gets under way, retailers, manufacturers, health care, and consumer durables tend to outperform. Lagging sectors include those tied to commodity prices, such as energy and basic industrial commodities, and also technology stocks.

Observations based on intermarket analysis can also help in allocating funds across national markets. Certain countries’ economies are closely tied to commodities—for example, Australia, Canada, and South Africa. As economies evolve, these relationships change. So, the relationships must be monitored closely. For example, the Chinese equity markets have become much more advanced since 2000, the Chinese economy is much more industrialized than in the past, and its dependence on exports is currently strong.

6. SUMMARY

• Technical analysis is a form of security analysis that uses price and volume market data, often graphically displayed.
• Technical analysis can be used for any freely traded security in the global market and is used on a wide range of financial instruments, such as equities, bonds, commodity futures, and currency futures.
• Technical analysis is the study of market trends or patterns and relies on recognition of patterns that have worked in the past in an attempt to predict future security prices. Technicians believe that market trends and patterns repeat themselves and are somewhat predictable because human behavior tends to repeat itself and is somewhat predictable.
• Another tenet of technical analysis is that the market brings together the collective wisdom of multiple participants, weights it according to the size of the trades they make, and allows analysts to understand this collective sentiment. Technical analysis relies on knowledgeable market participants putting this knowledge to work in the market and thereby influencing prices and volume.
• Technical analysis and fundamental analysis are equally useful and valid, but they approach the market in different ways. Technical analysis focuses solely on analyzing markets and the trading of financial instruments, whereas fundamental analysis is a much wider ranging field encompassing financial and economic analysis as well as analysis of societal and political trends.
• Technical analysis relies primarily on information gathered from market participants that is expressed through the interaction of price and volume. Fundamental analysis relies on information that is external to the market (e.g., economic data, company financial information) in an attempt to evaluate a security’s value relative to its current price.
• The usefulness of technical analysis is diminished by any constraints on the security being freely traded, by large outside manipulation of the market, and in illiquid markets.
• Charts provide information about past price behavior and provide a basis for inferences about likely future price behavior. Various types of charts can be useful in studying the markets: line charts, bar charts, candlestick charts, and point and figure charts.
• Relative strength analysis is based on the ratio of the prices of a security to a benchmark and is used to compare the performance of one asset with the performance of another asset.
• Many technicians consider volume information to be very important and watch for the confirmation in volume of a price trend or the divergence of volume from a price trend.
• The concept of trend is perhaps the most important aspect of technical analysis. An uptrend is defined as a security making higher highs and higher lows. To draw an uptrend line, a technician draws a line connecting the lows of the price chart. A downtrend is defined as a security making lower highs and lower lows. To draw a downtrend line, a technician draws a line connecting the highs of the price chart.
• Support is defined as a low price range in which the price stops declining because of buying activity. It is the opposite of resistance, which is a price range in which price stops rising because of selling activity.
• Chart patterns are formations appearing in price charts that create some type of recognizable shape.
• Reversal patterns signal the end of a trend. Common reversal patterns are the head and shoulders, the inverse head and shoulders, double tops and bottoms, and triple tops and bottoms.
• Continuation patterns indicate that a market trend in place prior to the pattern formation will continue once the pattern is completed. Common continuation patterns are triangles, rectangles, flags, and pennants.
• Price-based indicators incorporate information contained in market prices. Common price-based indicators are the moving average and Bollinger Bands.
• Momentum oscillator indicators are constructed from price data, but they are calculated so that they fluctuate either between a high and low, typically 0 and 100, or around 0 or 100. Some examples are momentum (or rate of change) oscillators, the RSI, stochastic measures, and MACD.
• Sentiment indicators attempt to gauge investor activity for signs of increasing bullishness or bearishness. Sentiment indicators come in two forms—investor polls and calculated statistical indices. Opinion polls to gauge investors’ sentiment toward the equity market are conducted by a variety of services. Commonly used calculated statistical indices are the put/call ratio, the VIX, margin debt, and the short interest ratio.
• Flow-of-funds indicators help technicians gauge potential changes in supply and demand for securities. Some commonly used indicators are the ARMS Index (also called the TRIN), margin debt (also a sentiment indicator), mutual fund cash positions, new equity issuance, and secondary equity offerings.
• Many technicians use various observed cycles to predict future movements in security prices; these cycles include Kondratieff waves, decennial patterns, and the U.S. presidential cycle.
• Elliott Wave Theory is an approach to market forecasting that assumes that markets form repetitive wave patterns, which are themselves composed of smaller and smaller subwaves. The relationships among wave heights are frequently Fibonacci ratios.
• Intermarket analysis is based on the principle that all markets are interrelated and influence each other. This approach involves the use of relative strength analysis for different groups of securities (e.g., stocks versus bonds, sectors in an economy, and securities from different countries) to make allocation decisions.

PROBLEMS

1. Technical analysis relies most importantly on:

A. Price and volume data.

B. Accurate financial statements.

C. Fundamental analysis to confirm conclusions.

2. Which of the following is not an assumption of technical analysis?

A. Security markets are efficient.

B. The security under analysis is freely traded.

C. Market trends and patterns tend to repeat themselves.

3. Drawbacks of technical analysis include which of the following?

A. It identifies changes in trends only after the fact.

B. Deviations from intrinsic value can persist for long periods.

C. It usually requires detailed knowledge of the financial instrument under analysis.

4. Why is technical analysis especially useful in the analysis of commodities and currencies?

A. Government regulators are more likely to intervene in these markets.

B. These types of securities display clearer trends than equities and bonds do.

C. Valuation models cannot be used to determine fundamental intrinsic value for these securities.

5. A daily bar chart provides:

A. A logarithmically scaled horizontal axis.

B. A horizontal axis that represents changes in price.

C. High and low prices during the day and the day’s opening and closing prices.

6. A candlestick chart is similar to a bar chart except that the candlestick chart:

A. Represents upward movements in price with X’s.

B. Also graphically shows the range of the period’s highs and lows.

C. Has a body that is light or dark depending on whether the security closed higher or lower than its open.

7. In analyzing a price chart, high or increasing volume most likely indicates which of the following?

A. Predicts a reversal in the price trend.

B. Predicts that a trendless period will follow.

C. Confirms a rising or declining trend in prices.

8. In constructing a chart, using a logarithmic scale on the vertical axis is likely to be most useful for which of the following applications?

A. The price of gold for the past 100 years.

B. The share price of a company over the past month.

C. Yields on 10-year U.S. Treasuries for the past five years.

9. A downtrend line is constructed by drawing a line connecting:

A. The lows of the price chart.

B. The highs of the price chart.

C. The highest high to the lowest low of the price chart.

10. The following exhibit depicts GreatWall Information Industry Co., Ltd., ordinary shares, traded on the Shenzhen Stock Exchange, for late 2008 through late 2009 in renminbi (RMB).

CANDLESTICK CHART GreatWall Information Industry Co., Ltd. Price Data, November 2008–September 2009 (price measured in RMB × 100)

Based on this chart, the uptrend was most likely broken at a level nearest to:

A. 7 RMB.

B. 8.5 RMB.

C. 10 RMB.

11. The “change in polarity” principle states which of the following?

A. Once an uptrend is broken, it becomes a downtrend.

B. Once a resistance level is breached, it becomes a support level.

C. The short-term moving average has crossed over the longer-term moving average.

12. The following exhibit depicts Barclays ordinary shares, traded on the London Stock Exchange, for 2009 in British pence.

CANDLESTICK CHART Barclays PLC Price Data, January 2009–January 2010 (price measured in British pence)

Based on this chart, Barclays appears to show resistance at a level nearest to:

A. 50p.

B. 275p.

C. 390p.

13. The following exhibit depicts Archer Daniels Midland Company common shares, traded on the New York Stock Exchange, for 1996 to 2001 in U.S. dollars.

CANDLESTICK CHART Archer Daniels Midland Company, February 1996–February 2001

This chart illustrates most clearly which type of pattern?

A. Triangle.

B. Triple top.

C. Head and shoulders.

14. In an inverted head and shoulders pattern, if the neckline is at €100, the shoulders are at €90, and the head is at €75, the price target is closest to which of the following?

A. €50.

B. €110.

C. €125.

15. Which flow-of-funds indicator is considered bearish for equities?

A. A large increase in the number of IPOs.

B. Higher-than-average cash balances in mutual funds.

C. An upturn in margin debt but one that is still below the long-term average.

16. A TRIN with a value of less than 1.0 indicates:

A. The market is in balance.

B. There is more volume in rising shares.

C. There is more volume in declining shares.

17. Bollinger Bands are constructed by plotting:

A. A MACD line and a signal line.

B. A moving-average line with an uptrend line above and downtrend line below.

C. A moving-average line with upper and lower lines that are at a set number of standard deviations apart.

18. Which of the following is not a momentum oscillator?

A. MACD.

B. Stochastic oscillator.

C. Bollinger Bands.

19. Which of the following is a continuation pattern?

A. Triangle.

B. Triple top.

C. Head and shoulders.

20. Which of the following is a reversal pattern?

A. Pennant.

B. Rectangle.

C. Double bottom.

21. Which of the following is generally true of the head and shoulders pattern?

A. Volume is important in interpreting the data.

B. The neckline, once breached, becomes a support level.

C. Head and shoulders patterns are generally followed by an uptrend in the security’s price.

22. Nikolai Kondratieff concluded in the 1920s that since the 1780s, Western economies have generally followed a cycle of how many years?

A. 18.

B. 54.

C. 76.

23. Based on the decennial pattern of cycles, how would the return of the Dow Jones Industrial Average (DJIA) in the year 2015 compare with the return in 2020?

A. The return would be better.

B. The return would be worse.

C. The answer cannot be determined because the theory does not apply to both of those years.

24. According to the U.S. presidential cycle theory, the DJIA has the best performance during which year?

A. The presidential election year itself.

B. The first year following a presidential election.

C. The third year following a presidential election.

25. What is a major problem with long-term cycle theories?

A. The sample size is small.

B. The data are usually hard to observe.

C. They occur over such a long period that they are difficult to discern.

26. In 1938, R. N. Elliott proposed a theory that equity markets move:

A. In stochastic waves.

B. In cycles following Fibonacci ratios.

C. In waves dependent on other securities.

27. All of the following are names of Elliott cycles except:

A. Presidential.

B. Supercycle.

C. Grand supercycle.

28. To identify intermarket relationships, technicians commonly use:

A. Stochastic oscillators.

B. Fibonacci ratios.

C. Relative strength analysis.

1Fundamental analysts use a wide variety of inputs, including financial statements, legal documents, economic data, first-hand observations from visiting the facilities of subject companies, and interviews with corporate managers, customers, suppliers, and competitors.

2Akerlof and Shiller (2009).

3The New York Society of Security Analysts was a successor to the New York Society of Financial Statisticians, which was founded in 1916.

4An American Depositary Receipt is a negotiable certificate issued by a depositary bank that represents ownership in a non-U.S. company’s deposited equity (i.e., equity held in custody by the depositary bank in the company’s home market).

5A European Depositary Receipt is a negotiable certificate issued by a depositary bank in one country against equity that is traded on the stock exchange of another country.

6This indicator is sometimes called the Wilder RSI.

7Brown (1999).

8This tool was first proposed by Richard W. Arms, Jr., a well-known technical analyst.

9In U.S. presidential election years, the vice presidency, all 435 House of Representatives seats, and 33 of the 100 Senate seats are also up for election.

10Breadth is defined as the ratio of the number of advancing securities in an index or traded on a given stock market to the number of declining issues.

11Murphy (1991).