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]


17

-4 is shown as a dotted hne. To this we will now add one more component - of still shorter duration and, of course, of correspondingly smaller amplitude. For the first time we will also assume the magnitude-duration fluctuation to be expected from the price-motion model. Therefore, to the elements that formed Figure III-4, we have added two additional ingredients:

1. A third cyclic component.

2. Magnitude-duration fluctuation-thus getting the simidation a little closer to agreement with our more complex model.

Once more we sum up each of the elements proposed to get the result at the top of Figure III-8. This time we have a pattern which abounds on all charts. The chartist variously refers to this type of formation as a "coil," "triangle," "flag," "pennant," "box," "diamond" or "wedge," All are formed from cyclic elements in the same manner. The differences are dependent upon the state of magnitude-duration fluctuation of a short duration component at the time when the sum of all longer duration components is pausing due to cyclicality. In each case, the rules of charting which unpute probabilistic follow-on to such patterns are based soundly upon the existence and behavior of the cyclic elements of our model.

HOW TO TELL IN ADVANCE IF A CHART PATTERN WILL "FAIL"

Triangles may be formed also in much the same manner as double tops and head-shoulder patterns. Lets examme in some detail a specific case which is reported in a popular book on charting. The stock is Perkin Elmer and the time period is March through September 1961. This time a daily high-low chart is used so that we can see the impact of shorter duration periodicities. Two triangles were shown in the book on chartmg in which this example appeared, and these are reproduced in Figure III-9.

Now we must examine this chart carefully in the light of our model. Two samples of a periodicity are shown with lows at A, B, and C. The average duration of these is 13.9 weeks, which we note is satisfyingly close to the 13-week nominal duration cycle of our model. Each of these is seen to consist of three shorter duration elements, the lows of which are shown on the figure as points 1 through 7. Six samples are present, averagmg 4.6 weeks. This is the 1961 time period equivalent of the nominal 3.25-week component of our model. Each of these consists of two shorter duration regularities. Twelve samples are present with an average duration of 2.3 weeks, again the 1961 equivalent of the 1.62S-week nominal component of our model. This situation is a good example of the deviations from model nominality that can be expected in individual issues and for varying periods of time. Each of the 2.3-week durations also shows a one-half duration cycle of 1.15 weeks average. This periodicity has not been mentioned in our model as yet, except for the statement that such shorter duration regularities exist.

Now inspect the small triangle at the top of the figure closely. consists of three cycles of the J. 15-week periodicity/ Magnitude variation is clearly evident here, with the result being more properly described as the chartists "flag." The breakout on the downside denoted trouble to the chartist-as well it should! Lets see what the cyclic situation was at this point in time:



Triangle Analysis In Peikln-Elmer

1. 13.9-weck cycle

2. 4.6-week cyde

3. 2.3-week cycle

4. 1.15-week cycle

8 weeks along-Aard down 3.8 weeks along-Aanf down 1.3 weeks along-and hard down 1,1 weeks along-and hard down

Is it any wonder that a downside breakout of the triangle occurred?

Actually, something else of interest should also be noted here. The charting rule for such triangles is: The probabilities favor the continuation of the trend that preceded the triangle formation. As pointed out in the book from which this example was taken, this triangle represented a failure of charting expectations. No reasons for such failures are usually postulated, but we can easily see how our new knowledge of cyclicality allowed us to forecast this failure from a triangle! Such a capability adds immensely to the utility of such chart patterns. Not only can we understand how and why they are formed (so they are no longer a mystery to us), but we can often read much more into them than can the conventional chartist!

Now lets look at the large triangle to the right of the chart. This one is formed in a completely different way than the preceding one. The triangle is composed of one and most of a second cycle of our 4.6-week periodicity. The 13.9-week cycle is going over a top, and the same formation mechanics occur as in the case of double tops and head-shoulder formations. It is only the particular combination of magnitude-dominant cyclic durations, the time relationships between these, and the particular rate at which the sum of all longer duration components is moving that make the formation appear as it is.



If we were watching this triangle fonn, how would we analyze the situation? The chartist would apply his probabilistic rule and say he expects a downside breakout - and he would be right. Would we come up with the same conclusion?

1. 13.9-week cycle 7.4 weeks along-Zuwd down

2. 4.6-WBek cycle 3.6 weeks along-Aaraf down

3. 2.3-week cycle 13 -weeks along-ftoref ddvwi

4. 1.15-week cycle bottoming oMt-fiat to up

In addition to this we note that the preceding 13.9-week cycle has formed a lower low. This tells us that the sum of all components with durations In excess of 13.9 weeks is also hard down. Under such circumstances we would positively predict a downside breakout, and again be right! We see that our cyclic model has enabled us not only to agree with the chartist when he is probabilistically correct, but also to (correctly) disagree with him when he is wrong. In short, we can explain why such patterns form, why they provide more right answers than wrong ones, and in addition, reduce the number of times in which a wrong decision is made! We begin to see some of the utility of our "X motivation" model.

The chartists patterns and our price model work hand-in-hand. The chart patterns are an aid in calling our attention to specific cyclic activity. The chartists expectations for the pattern may then be verified or repudiated, based upon cyclic considerations.

UNDERSTANDING OTHER CHART PATTERNS

To carry our comparisons further, let us now consider "V" top and bottom patterns. These occur when an intermediate duration cycle tops or bottoms out in perfect time synchronization with the next longer duration component. The "V" is further accentuated when the next shorter duration cycle also tops or bottoms simultaneously. In cases where four or five of these behave in this manner, the "V" can be very exaggerated and sharp indeed!

Saucer formations, measured moves, gaps, islands, one-day reversals, and all other well known chart patterns are as completely identifiable and explainable by cyclic theory as are the patterns discussed. It would be a very good exercise for you, the reader, to go through each of these situations, and prove this statement for yourself

HOW CYCLICALITY GIVES MEANING TO CHART PATTERNS

But lets move on to other areas of mterest. One of the biggest difficulties in charting to date consists of the inability of the chartist to answer the following question: "How do you know which fluctuations (separated by how much time) to use in setting up your chart patterns?"

We have already seen in preceding chapters that an envelope may be dravra about any and all of our cyclic components. Just so, when a chartist draws a straight segment of a channel or trend line (representing part of our cyclic channel), we may always find other fluctuation lows or highs which define other channels or trend lines during the same time period and for the same issue. Which of these has valid meaning? Certainly,



[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]