Technical Analysis Explained

by Martin J Pring

  The Arms Index can also be used with a moving average where the 10-day (open TRIN) time span is the most widely followed. It is interpreted in much the same way as the 10-day A/D ratio discussed earlier in this chapter. Most of the time, these two series move in a consistent manner, but from time to time the Arms Index gives some subtle indications that the prevailing trend is about to reverse. Generally speaking, when the 10-day Arms rises above 150, this signals a major low. Sometimes this happens right away; at other times there is a delay of 10 to 20 days before the final bottom is seen. Between 1968 and 2001 there were no exceptions to this rule. Chart 26.22 shows three different versions of the Arms Index with a 10-, 25-, and 45-day time span. The rationale for including three indicators in the chart was explained in Chapter 13. The arrows indicate those points when at least two of these inversely plotted series are at an extreme and have begun to reverse, since tops tend to be rolling affairs where volume leads price. The Arms Index often reaches its extreme ahead of the market average, whereas selling climax conditions seen at bottoms are usually far more coincident in nature.

  CHART 26.22 S&P Composite, 1997–2001, and Three Arms Indicators

  On Balance Volume

  On Balance Volume (OBV) was discovered by Joe Granville and published in his book Granville’s New Key to Stock Market Profits (Literary Licensing, 2011). The indicator is plotted as a continuous, cumulative line. It begins with an arbitrary number, which rises and falls depending on what the price does. The volume for the day is added in when the price rises, and is subtracted when it falls. If intraday charts are being used, volume units would be added and subtracted based on the time frame of the bars. For weekly charts, the basis would be the week and so forth. OBV, therefore, offers a rough approximation for buying and selling pressure and has become a very popular indicator. It is interpreted by comparing the line to the price, using divergences, trendline breaks, price patterns, and MA crossovers to point up underlying strengths or weaknesses.

  Chart 26.23, featuring RF Micro Devices, shows a few OBV characteristics. We see a negative divergence right at the beginning of 2011 and this is later confirmed at A, where both series make new lows. A positive divergence, where the price makes a new low but the OBV does not, develops at B, but it was not possible to observe any joint trendline breaks because the action was too volatile. Finally, at C it was possible to construct two trendlines, both of which were eventually violated.

  CHART 26.23 RF Micro, 2010–2011, and an On Balance Volume Line

  I do not find the OBV indicator to be as accurate as this in most situations. Indeed, its warnings are often as misleading as the valid signals. My advice, therefore, is to tread carefully with this indicator and make sure its signals are backed up with other approaches. This is especially true for positive and negative divergences, but not so much with the joint trendline break, which I find to be more accurate. Chart 26.24 featuring Alergan, for instance, shows a situation at the end of 1999 when OBV was pointing to higher prices, but they went down; in early 2000, weaknesses in the OBV pointed to lower prices, but they went up! Note that in both charts the joint trendline technique worked, which as mentioned earlier, is probably the best way to interpret this indicator.

  CHART 26.24 Alergan, 1998–2000, and OBV

  Equivolume

  Equivolume is a plotting concept developed by Dick Arms (www.armsinsider.com). It is similar to the candlestick volume approach discussed earlier. Bars are plotted in different widths depending on the level of volume for that particular period. The greater the volume, the wider the bar. The top and bottom of the bar represent the high and low for that period. This is a very useful approach because it graphically shows in one series whether prices are rising or falling on light or heavy volume. The width of the box is controlled by a normalized volume value. The volume for an individual box is normalized by dividing the actual volume for the period by the total of all volume displayed on the chart. Therefore, the width of each equivolume box is based on a percentage of total volume, with the total of all percentages equaling 100.

  The resulting charts represent an important departure from all other analytical methods, in that time becomes less important than volume in analyzing price moves. It suggests that each movement is a function of the number of shares or contracts changing hands rather than the amount of time elapsed.

  Because of this, the dates on the x axis are not equidistant from each other, as is normally the case. They instead depend on the volume patterns for their location. Chart 26.25 shows an equivolume arrangement for MMM Company. At A the price breaks out with a couple of thick bars, indicating very heavy volume. In effect, this is a classic buy signal. At B the rally is associated with very narrow equivolume bars, which tells us that there is a distinct lack of upside volume. Thus, a warning of an impending trend reversal is given.

  CHART 26.25 MMM, 2000–2001, and Equivolume

  In Chart 26.26 for Aditya Birla Nuvo, you can see the trading range in November 2007 is associated with extremely low volume (narrow bars).

  CHART 26.26 Aditya Birla Nuvo, 2007–2008, and Equivolume

  Then the price breaks to the upside with a wider (high-volume bar). Just after the January 2008 peak the bars widen again and the falling price/expanding volume combination results in a sharp drop. The tail end of the May and August rallies is also accompanied by a bearish shrinkage in volume. Note how the August rally runs into resistance at the two previous very wide bars at the two dashed lines at X. At the August high, nearly anyone who had bought since April was coming home with a loss, and that meant some oversupply pressuring the price. When the price fell to the line at Y, it took it below the heavy volume bar in July, thereby adding even more unhappy traders to the already high total.

  Summary

  1. The ROC of volume often gives signs of subtle changes in the level of activity that are not apparent from volume data represented as a histogram.

  2. The ROC of volume can be expressed as a percentage or in a subtraction format.

  3. Volume ROCs and oscillators can be used with overbought/oversold crossovers, trendline analysis, and price patterns.

  4. Overbought readings in the volume ROC can be followed by declining or rising prices, depending on the nature of the previous trend.

  5. The Demand Index is constructed from volume and price; moves in the same direction as a regular price oscillator; and is best used with divergence and overbought/oversold analysis, trendline, and price pattern construction.

  6. The Chaikin Money Flow is constructed from volume and price data and moves in the same direction as a regular price oscillator. It is best used with divergence analysis.

  7. Upside/downside volume measures the volume in advancing and declining stocks. It can be used as a continuous line or in oscillator format.

  8. The Arms Index is constructed from advancing and declining stocks and their respective volume. It is usually calculated over a 10-day span. Readings in excess of 150 signal major bottoms.

  9. OBV is constructed as a continuous line and used with divergence analysis. Joint trendline breaks between the OBV line and the price offer a more accurate method of interpretation.

  10. Equivolume is a form of charting that plots the thickness of each bar according to the volume experienced during its formation. The thicker the bar, the higher the level of activity and vice versa.

  27 MARKET BREADTH

  Breadth indicators measure the degree to which the vast majority of is-sues are participating in a market move and, therefore, monitor the ex-tent of a market trend. Generally speaking, the fewer the number of issues that are moving in the direction of the major averages, the greater the probability of an imminent reversal in trend. Breadth indicators were originally developed to monitor trends in the stock market, but in recent years, they have been expanded to embrace any market that can conveniently be subdivided into components. Even though most of the comments in this chapter refer to U.S. equities, it should be remembered that breadth
can just as validly be extended to other markets around the world. For example, experiments with Indian, Brazilian, and several Middle Eastern markets have shown they are susceptible to such analysis, and I see no reason why other countries would not also lend themselves to this approach.

  Breadth analysis can be applied to any sector or market that can be broken down into a basket of securities that are reflective of an overall index. An example might include a selection of commodities being compared to a commodity index or a series of currencies to an overall currency index such as the Dollar Index and so forth. The main thing to bear in mind is that the principles of interpretation remain constant.

  The concept of breadth can probably be best explained using a military analogy. In Figure 27.1, lines AA and BB indicate military lines of defense drawn up during a battle. It might be possible for a few units to cross over from AA to BB, but the chances are that the BB line will hold unless an all-out effort is made. In example a, the two units represented by the arrows are quickly repulsed. In example b, on the other hand, the assault is successful since many units are taking part, and army B is forced to retreat to a new line of defense at B1.

  FIGURE 27.1 Trench Warfare

  A narrowly advancing stock market can be compared to example a, where it looks initially as though the move through the line of defense (in stock market terms, a resistance level) is going to be successful, but because the move is accompanied by such little support, the overall price trend is soon reversed. In the military analogy, even if the two units had successfully assaulted the BB defense, it would not be long before army B would have overpowered them, for the farther they advanced without broad support, the more vulnerable they would have become to a counteroffensive by army B.

  The same is true of the stock market, for the longer a price trend is maintained without a follow-up by the broad market, the more vulnerable is the advance.

  At market bottoms, breadth is not such a useful concept for determining reversals because the majority of stocks usually coincide with or lag behind the major indexes. On the few occasions when breadth reverses its downtrend before the averages, it is actually a more reliable indicator than at tops. I’ll begin this discussion with a rationale as to why the broad market normally leads the averages at market tops. The word “normally” is used because, in the vast majority of cases, the broad list of stocks does peak out ahead of a market average such as the Dow Jones Industrial Average (DJIA) or the S&P Composite. This rule is not invariable, however, and it should not be assumed that the technical structure is necessarily sound just because market breadth is strong. In most cases it will be, but if other indicators are pointing up weakness, this can override a positive breadth picture.

  Advance/Decline Line

  The most widely used indicator of market breadth is an advance/decline (A/D) line. It is constructed by taking a cumulative total of the difference (plurality) between the number of New York Stock Exchange (NYSE) issues that are advancing over those that are declining in a particular period (usually a day or a week). Similar indexes may be constructed for the American Exchange (AMEX) or NASDAQ issues. Because the number of issues listed on the NYSE has expanded since breadth records were first kept, an A/D line constructed from a simple plurality of advancing over declining issues gives a greater weighting to more recent years. For the purpose of long-term comparisons, it is better to take a ratio of advances versus declines, or a ratio of advances and declines divided by the number of unchanged issues, rather than limiting the calculation to a simple plurality.

  The late Hamilton Bolton devised one of the most useful measurements of breadth. It is calculated from a cumulative running total of the following formula:

  where A = the number of stocks advancing

  D = the number declining

  U = the number unchanged

  Since it is not mathematically possible to calculate a square root of a negative answer (i.e., when the number of declining stocks is greater than the number of those advancing), the D and A are reversed in such cases, so that the formula becomes the square root of D/U – A/U. The resulting answer is then subtracted from the cumulative total, as opposed to the answer in the earlier formula, which is added. Table 27.1 illustrates this calculation using weekly data.

  TABLE 27.1 Weekly A/D Line Calculation (Bolton Formula)

  Inclusion of the number of unchanged issues is useful because the more dynamic the move in either direction, the greater the tendency for the number of unchanged stocks to diminish. Consequently, by giving some weight to the number of unchanged stocks in the formula, it is possible to assess a slowdown in momentum of the A/D line at an earlier date, since an expanding number of unchanged issues will have the tendency to re-strain extreme movements.

  The A/D line normally rises and falls in sympathy with the major market averages, but it usually peaks well ahead of them. There appear to be three basic reasons why this is so:

  1. The market as a whole discounts the business cycle and normally reaches its bull market peak 6 to 9 months before the economy tops out. Since the peak in business activity is itself preceded by a deterioration of certain leading sectors such as financial, consumer spending, and construction, it is logical to expect that the stocks representing these sectors will also peak out ahead of the general market.

  2. Many of the stocks listed on the NYSE, such as preferreds and utilities, are sensitive to changes in interest rates. Since interest rates usually begin to rise before the market peaks, it is natural for the interest-sensitive issues to move down in sympathy with rising rates.

  3. Poorer-quality stocks offer the largest upside potential, but they are also representative of smaller, underfinanced, and badly managed companies that are more vulnerable to reduced earnings (and even bankruptcy) during a recession. Blue-chips normally have good credit ratings, reasonable yields, and sound underlying assets; thus, they are typically the last stocks to be sold by investors during a bull market.

  The DJIA and other market averages are almost wholly composed of larger companies, which are normally in better financial shape. These popular averages, therefore, continue to advance well after the broad market has peaked out.

  Interpretation

  Here are some key points for interpreting A/D data:

  1. Some A/D lines appear to have a permanent downward bias. It is, therefore, important as a first step to observe the relationship between an A/D line and an index over a very long period to see whether this bias exists. Examples include breadth data for the AMEX market, the U.S. over-the-counter (OTC) market, and the Japanese market.

  2. Divergences between a market average and an A/D line at market tops are almost always cleared up by a decline in the average. However, it is mandatory to await some kind of trend-reversal signal in the average as confirmation before concluding that it will also decline.

  3. It is normal for the A/D line to coincide or lag at market bottoms. Such action is of no forecasting value. When the A/D line refuses to confirm a new low in the index, the signal is unusual and very positive, but only when confirmed by a reversal in the average itself.

  4. Breadth data may diverge negatively from the averages, but an important rally is often signaled when a down trendline violation is signaled along with a breakout in the market average itself.

  5. In most cases, daily A/D lines have more of a downward bias than lines constructed from weekly data.

  6. A/D lines may be used with moving-average (MA) crossovers, trendline breaks, and price pattern analysis. For longer periods, the 200-day MA appears to work reasonably well.

  7. When the A/D line is in a positive trend, e.g., above its 200-day MA, it indicates that the environment for equities in general is a positive one, regardless of what the major averages such as the DJIA or S&P Composite may be doing. A positive A/D line is, therefore, a better bellwether for the market as a whole than a narrowly based blue-chip index. The opposite is true when the A/D line is in a declining trend.

  Ma
jor Technical Principle The longer and greater the negative divergence between the A/D line and the market average it is monitoring, the deeper and more substantial the implied decline is likely to be.

  For this reason, divergences between the A/D line and the major market averages at primary peaks are more significant than those that occur at intermediate tops. For example, Chart 27.1 shows that the weekly A/D line peaked in March 1971, almost 2 years ahead of the DJIA, a very long period by historical standards. The ensuing bear market was the most severe since the Depres-sion. On the other hand, the absence of a divergence does not necessarily mean that a steep bear market cannot take place, as the experience of the December 1968 top indicates. This is also shown in Chart 27.1.

  CHART 27.1 The DJIA and the Weekly NYSE A/D Line, 1966–1977

  Positive divergences develop at market bottoms where the A/D line refuses to confirm a new low in the Dow. The most significant one occurred in the 1939–1942 period. The DJIA (shown in Chart 27.2) made a series of lower peaks and troughs between 1939 and 1941, while the A/D line refused to confirm. Finally in the middle of 1941, the A/D line made a post-1932 recovery high unaccompanied by the DJIA. The immediate result of this discrepancy was a sharp sell-off into the spring of 1942 by both indicators, but even then the A/D line held well above its 1938 bottom, unlike the DJIA. The final low in April 1942 was followed by the best (in terms of breadth) bull market on record. This positive action by the broad market is unusual. Typically at market bottoms the A/D line either coincides with or lags behind the low in the DJIA and has no forecasting significance until a reversal in its downtrend is signaled by a breakout from a price pattern, a trendline, or MA crossover.