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]


20

chapter four

Timing Your Buys With Graphics

• Prediction by Graphics

• How to Construct Curvilinear Envelopes

• Prediction of Price Turns Using Envelopes

» Use This Example as Your Channel Prediction Guide

• Constructing the Dominant Channel

• Finding the Outer Envelope

• Setting Up Price-Turn Predictions

• The Kind of Results You Can Achieve

• How to Generate Grafic "Buy" Signals

• What to Look For

• Recognizing the "Valid Trend Line"

• "Edge-Band" Transaction Timing

• "Mid-Band" Transaction Timing

• Points to Remember Regarding Graphic "Buy" Signals

PREDICTION BY GRAPHICS

You have seen what is required to maximize investment profits. Transaction timing is the golden key-and the price-motion model implies a way. However, concepts of the price-motion model must now be applied in a practical manner to generate objective investment decisions. Of all the methods available to do this, graphical techniques are by far the fastest and easiest to apply. This chapter will tell you hoW to form and interpret curviHnear envelopes, and how to combine price-motion model concepts with conventional chart patterns to provide definitive "wait," "buy." and "hold" signals.



The first and most basic step is the identification of the dominant cyclic components present in a particular stock, and the status of these components at the present time. The objective is what we might call "first-order" prediction. That is, we want to derive transaction timing benefits from the price-motion model by giving ourselves a high statistical probability of correctly estimating when a set of cyclic components wil! form a low (or a high). We must use knowledge of cyclicality coupled with past price action in order to form an opinion of probable price turning points in the ftiture-and we wish to do this using simple and fast graphical techniques.

HOW TO CONSTRUCT CURVIUNEAR ENVELOPES

The constant-width envelope that you were introduced to briefly in Chapter Two is a powerful tool for the accomplishment of this task. You have seen several examples of such envelope analysis now. In each case data on the stock of interest was presented in a figure before addition of envelope bounds. This was done to provide a feeling for the large increase in visibility afforded by the technique. Chartmg a stock tells you much more at a glance than looking at tabulated data. Similariy, addition of constant-width envelopes brings out and stresses the cyclic price-time relationships on the chart. It would be a good idea at this point for you to return to and study the pairs of such figures presented in previous chapters.

To start, you must construct or procure a weekly high-low chart of the issue you are interested in. The time period covered by the chart should be at least long enough to include one and a half cycles of the periodicity that is next longer in duration than the one you are mterested in trading on. If practical, it is even deshable to include the second longer duration component in your data. Remember that these long-term regularities have much larger magnitudes than your trading cycle, and can therefore be very mfluential in determining where the price of the stock goes, regardless of what your trading cycle is doing!

By observation of your weekly chart, you will always find that some one cyclic component is most clearly discernible. This will usually be some variation of the 13- or 26-week (nominal) periodicity of the model. Keep the general outline of this price motion ui mind while you look for a narrow band of prices which will essentially enclose the data, and will fluctuate with the dominant movement.

Start by lightly sketching in upper and lower bounds to this band (an overiay of tissue or vellum is of help here). Now use a piece of scratch paper to tnensure the channel heigiit vertically at several points. Choose an average of these and correct your embryonic envelope to width, if any data points cannot be included in

this channel, widen the envelope until all price motion is accounted for.

Now repeat the process while decreasing the width of the band. You will normally find a minimum-width envelope which contains all of the data except two or three stray points which will be obvious by virtue of being the only data in the space between the two envelopes youve drawn. Erase the wider of the two envelopes, for it has now served its purpose.

The final envelope will now enable you to make a more accurate estimate of the



duration of the dominant periodicity by clearly demaridng locations of highs and lows, free from distortion caused by the presence of fluctuations of shorter duration. Tabulate your estimate.

Now look within the envelope youve drawn. Note the data touches or comes close to envelope bounds. Price motion will tend to sweep back and forth between these limits-somewhat regulnr p;-- " - io<,t clesr -r-rnent and

measure between the lows, again using marks on scratch paper. Look now for other lows both forward and backward in time from the first one. Knowing in this way the approximate location at which you should find such additional lows, it is usually apparent where they are, even when duration-magnitude fluctuation is present. As each one is located, mark it on the chart. Note the dir each ?;imple fv..... ,;J averac? these. This is yor.!- best estimate of the expected duration of the component contained v.ithin your first envelope. A second envelope can so!7ietimes be constructed about this component-but more often than not you will find it to be too irregular to be of much help.

Go back to your original channel and examine it closely. If you have charted sufficient data you wUl have at least two highs and two lows of the dominant component shown. These peaks and valleys constitute points on a second envelope which contains the dominant component. Sketch this one in, using the same techniques as for the first. You have now identified three of the cyclic components dominant in the specific issue you have chosen!

With this much accompUshed, you can normally spot still shorter duration periodicities within your narrowest channel. These will typicaUy display too much variation in duration and will be too near the next longer duration component in size to allow identification using other envelopes. However, you can mark the lows involved with a check or something similar.

PREDICTION OF PRICE TURNS USING ENVELOPES

Prepare yourself a table showing the condition of each of the identified components just prior to each major up or down move on your chart. Do this by using the average durations you have obtained for the samples present of each cycUcality. Then count weeks from the last low preceding the point of interest. Remember, any given periodicity reaches a peak and starts back down one-hatf of the average duration from the last low. You wUl find that, if you have property drawn your channels, the resulting tabulations wUl completely explain every fluctuation you see in the stock price!

Now, repeat your tabulation of cyclic condition for the last data point on your chart. What you are looking for, of course, is the case where two or more cyclic components are due to bottom out ui the very near future, at the same time, and the sum of all longer duration components is hard up. Once found, you are ready to track the « closely, determined to buy the moment price action tells you the process you expect has started! If you use your average cychc durations and neasure from the last visible low of each cycle into the future, you will have a pretty fair idea as to just how soon your ideal buy point may occur.

A few comments on the above. You may wish at this time to change your mind



[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]