back start next


[start] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [ 29 ] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72]


29

Dov, Prices

- 2900

- 2800

- 2700

2600

2500

2400

2300

7/23

8/15

9/15

10/1

Figure 8.12 The 1990 Dow Industrials, with my long-term price oscillator. This long-term price oscillator has worked well for me since I first began using it in 1975. Note how the oscillator turns up before the market did in late September.

of some important market parameter such as price, breadth, moving averages, or v olume; and they oscillate around a baseline, much like the plot of the momentum of the pendulum.

For the equities markets, I use two oscillators-one for breadth and one for price. For commodities, I use an oscillator based upon the difference between two moving averages.

The oscillators I use for the stock indexes, price and breadth, have remained valid since I first started using them in January 1975" (see Figures 8.12 and 8.13). The one I use for commodities is built into my quote system, and you can change the period of the moving averages until you find "a good fit" for the particular commodity you are dealing with as in Figure 8.14.

Refeting to the figures, you can see that a good oscillator begins reversing when or before the market trend begins to change. In most cases, these particular oscillators are quite accurate. Typically, when the market is in an uptrend, the

Actually, I did use two price oscillators for the Industrials, one short-term and one longterm, but the advent of program trading in the mid-80s invalidated the shorter-term oscillator.



- 2400

- 2300

Figure 8.13 The 1990 Dow Industrials, with my breadth oscillators. The lighter line is the short-term breadth oscillator, and the darker one the long-term. Compare the movement of these oscillators with the movement of the Down in Figure 8.12.

oscillator will also be moving up. The higher the oscillator rises on the chart, the more the market approaches an "overbought" condition. Ideally, just like the plot of the pendulum, as the market approaches the top, the oscillator will "slow down," that is, flatten out and top. Then, as prices start to reverse, the oscillator will tum down and continue down until the market approaches an "ove rsold" condition, when the reverse of what happens at a top occurs.

Generally speaking, the higher or lower the oscillator lines go above or below the baseline relative to other peaks and valleys in the plot, the more significant they become. Now let me sh ow you how to calculate the specific oscillators that I use and demonstrate exactly how I use them.

The best momentum measurement I use for stock market breadth is really what you might call a 10-day equivalent, net change, moving average, breadth oscillator. It sounds complicated, but it is really quite simple. Before I get into it, however, let me say a few words about breadth..

Breadth is an indicator used to counterbalance the fact that most s tock average indices are weighted. The Dow Industrials, for example, is an average of only 30 stocks weighted to price. Sometimes, if a heavily weighted stock makes an unusu



Ddily Bar Chart

Figure 8.14 Daily bar chart of Cattle Futures, with overboughoversold oscillator. The oscillator on the bottom is based on the difference between the 7day and 21-day moving average. While this is a good fit for feeder cattle, it may not work well at all for other commodities.

ally extensive move, it can throw the average off as a good indicator of nationwide industrial stock performance. One way to test the validity of the averages as general market indicators is to compare them to what is called the advance/decline line (the A/D line), that is, breadth.

The A/D line is simply a plot of difference between the total number of advancing issues and the total number of declining issues of every stock on the New York Stock Exchange versus time in days. Generally, the AID line moves with the Dow, and when there is a divergence between them, it often signals a coming change in the trend.

The breadth oscillator is simply a measure of market momentum, and it often gives you an earlier signal of a coming change in trend than does the A/D line by itself.

I actually use two breadth oscillators, one short term, and one longer term, but I generally pay much more attention to the longer term one. For the short term, every morning, first I log the net A/D number; I simply calculate, log, and plot a moving sum of advancing issues minus declining issues on the NYFE for the previous 10 days. In other words, every day, I keep a running, cumulative total of the daily difference of advancing issues minus declining issues. The process is shown in more detail in Table 8.1.



[start] [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17] [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [ 29 ] [30] [31] [32] [33] [34] [35] [36] [37] [38] [39] [40] [41] [42] [43] [44] [45] [46] [47] [48] [49] [50] [51] [52] [53] [54] [55] [56] [57] [58] [59] [60] [61] [62] [63] [64] [65] [66] [67] [68] [69] [70] [71] [72]