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]


45

Figure IX-1 is a block diagram of the anatomy of price change. Into the box labeled "Conversion Mechanism" we will put all processes between an investor decision to buy or to sell and the consummation of the resulting trade. Stock price changes are considered as an "output" of this box, such chan being associated with each transacticm. The processes within the box we will now proceed to ignore, since they are the direct but not the elemental cause of price fluctuations. It is the decision to act that starts the whole process off.

But what causes a human bemg to come to a decision? Logic? Emotion? The press of circumstances?

All of these and much, much more. It should be clearly understood at this point that the actual processes by which decisions are made are not understood-but the more that is learned on this subject, the more complex the process appears. Our particular need is to simplify the process in our own thinking to such an extent that we can logically accept the strange new facts on which we now mtend to operate, even if we cannot fuUy understand what causes them to exist.

In Figure IX-1 such simplification is effected by lumping all possible contributors to the investment decision process under three headings:

1. Fundamental Factors

We know for a fact that people do buy and sell on the basis of research and knowledge of these quantities. Therefore we must assimie that such factors do influence investment decisions (hence price change) to some extent.

2. Random Factors

We also know that investors sometimes buy and sell stocks for no better reason than to place excess funds or to raise cash. Such independent and non-price-level-correlated activity is most likely to create some level of random "noise" in price action.

"X" Motivation

Now we simplify stiU more. For here we lump all other possible contributions, known or unknown, to the investment decision process-of any kind whatsoever.

Now we have an interesting handle on the situation. Consider:

1. You have seen examples in previous chapters of regular periodicity in price motion.

2. The nature and existence of such regularities can be described in detail. Techniques can be set Up which can only work if such regularities exist-and can be shown to work.

3. The nature and existence of such regularities can be proven conclusively by analysis through some of the methods and results of Chapter Eleven and the Appendbc.

4. But-and this is important-by definition, random price action (hence decision factors) cannot contribute to such regularity. Furthermore, to accept fundamental factors as the cause would force acceptance of the occurrence of fundamental events on a regular schedule, like a freight train, and we can readily show that this is not the case.

5. The conclusion is unavoidable that the cyclic regularities noted must be due to the lumped sum of all other possible contributors to human decision processes-or to what we have called "X" motivation!



UNDERSTANDING IRRATIONAL DECISION PROCESSES

First of all lets consider a hypothetical situation which has probably been experienced by every investor in one form or another.

Assume you have bought a stock on a fundamental tip. You are showing a paper profit and trymg to make a decision as to whether you should sell or not. Lets say the fundamental factor is still in effect and the principal reason you are considering selling is the profit you would show if you did. What other factors might enter into your thinking at this time?

Perhaps you have been unsuccessful in your last several transactions and need the "ego boost" of a success. Your wife may be getting impatient, and you could fear a curtailment of your investment activities in favor of, say, redecorating the house if you dont show black in your ledger soon. Youve seen paper profits dwindle to real dollar and cents loss before and you dont want it to happen again. Perhaps the automobile just broke down and you need the money anyway.

In making your decision, how can you possibly weigh and evaluate all of these considerations so as to arrive at a logical and rational decision? Obviously you cannot. So the resulting decision must be irrational in a sense!

Now lets add a purely emotional factor. Assume youve been giving the problem consideration, but have not made up your mind. One morning you awaken with a headache, the hot water is off, and youve got a flat tire. By the time you get to the office youre in a testy frame of mind. About mid-morning you suddenly pick up the phone and tell your broker to sell. What finalized your decision for you? It could be that with everything going awry that day you simply wanted at least one situation to be resolved favorably-in this case by a profit on your transaction. Suddenly your whole day seems brighter and you congratulate yourself on a good decision.

But supposing the next day the stock skyrockets in a short squeeze. You find you could have doubled your profits by waiting an additional two days. Now how do you feel regarding the validity of your decision?

This is a surprisingly good example of the way in which many decisions are made in the al»ence of sufficient facts on which to base a logical conclusion. Remove the facetiousness, rearrange the factors and circumstances a bit, and you could easily be describing many management decisions made by corporate executives. In fact, the

And this is the concept that is difficult to accept. To do so, we must admit the possibility that something causes millions of investors operating from widely differing locations, making countless buy and sell decisions, at varying points in time, to behave more or less alike-and to do so consistently and persistently! How can this be?

The answer to this question is not known, although reasonable theories can be formulated. The best that can be done here is to show that many things do influence decisions-often without conscious awareness of the fact-and that the unknowns in the decision process can conceivably account for the startUng behaviour of price fluctuations.



ability to make good decisions in the absence of sufficient infonnation is what distinguishes the successful executive (or investor) from the unsuccessful one!

So decisions are often based on other than a straightforward line of reasoning. Lets complicate the matter still more.

Recent experimental evidence tends to show that such intangibles as fatigue and frame of mind are mfluenced by the presence or absence of physical force fields. For example, Dr. Cristjo Cristofv (father of the "Cristofv Effect" used to detect nuclear explosions) reports that pilots of ll-2s and truck drivers show increased alertness and job efficiency when the Earths magnetic field, modified by the metal of the vehicle, is artificially restored. Now all of us exist in an environment that is sfanply riddled with force fields-gravitational, electrostatic, etc. If such fields can influence some physical and mental functions, might they not influence o/Aers-perhaps causing masses of humans to feel simultaneously bullish or bearish in the market, for example? Conjecture, but a possibility.

It is not being suggested here that the cyclic sub-model is based on this type of thing. It is enough to show that the possibility exists of external and unknown influence on investment decisions. Seeing how it might come about helps us to accept the evidence of our eyes and analyses, even when the results seem irrational by past experience.

WHAT YOU SHOULD KNOW ABOUT FUNDAMENTAL FACTORS

The discussions of the previous sections should help you to accept and apply the non-traditional concepts presented in eariier chapters with increased confidence.

In like vein, lets go back to several statements in Chapter Two which are a part of the price-motion model, but were presented without substantiation. One of these was: national and world historical events contribute in an utterly negligible way to the performance of the market as a whole, and that of individual issues in particular.

That this is a foreign and alien concept is amply illustrated by every financial column and commentary you read and hear. The market is constantly being cited as weak or strong depending on the progress of Vietnam peace talks or ten dozen other such factors.

Fortunately, its not too hard to check on just how much such things influence the market. Weve plenty of recorded market price history on which to draw, and the Almanac supplies corresponding dates of such events. Lets look into this one, because if you think such thin do influence the market you cannot help letting it influence your decisions. And if they do not, you are quite likely to make wrong decisions with attendant loss.

It was also stated without amplification that company-oriented fundamental factors contribute heavily to the broad, smooth sweep of price motion. Let us investigate all of these areas simultaneously.

There are certainly at least three broad classifications involved.

1. Factors related solely or primarily to a given company and its operations. New products, new management policie.s, and key personnel changes serve as examples.



[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]