The science of the stock market: is it possible to successfully predict on history?
The question of whether we can rely on past price performance data to predict future price performance arises very often. Frankly speaking, why it arises so often is not quite clear, because any forecasting methodFrom the weather forecast to the fundamental analysisIt is the data from the past that are relied upon. What else is there to rely on?
In the field of statistics, there is a clear distinction between descriptive statistics and inductive statistics. Descriptive statistics is concerned with the graphical display of data, so it includes, for example, the traditional bar graphs of prices. Inductive statistics is concerned with generalization, prediction and extrapolation, based on the information that can be drawn from descriptive statistics. Consequently, price charts themselves fall into the category of descriptive statistics, and the kind of analysis that technical analyst, - inductive.
Trust the technical analysts - proven
As one book on statistics says, "the first step in business or economic forecasting is to gather information about the past. (See John B. Freund and Frank Williamet, Modern Business Statistics, 1969, p.383.) Thus, chart analysis is nothing more than a form of time-sequence analysis based on the study of past data. The only information available to any analyst is information about the past. Only by projecting the facts of the past into the future can we evaluate the future. Here is another quote from the same source: "Demographic forecasting, business forecasting and all the like are largely based on facts that have already taken place. In business and science, in our daily lives, we are constantly relying on our past experiences to anticipate what is in store for us in an as yet unclear future.
Thus, the use of technical analysts statistical data on price dynamics in the past in order to forecast the future is quite valid and has a solid scientific basis. Anyone who would doubt the validity of such technical forecasts might as well doubt the validity of any forecasting data based on historical facts, including all economic and fundamental analysis.
The theory of "random events"
Random Walk Theory has a strictly scientific origin. Its basic premise is that price fluctuations are "serially independent," so past price data cannot be used to make reliable predictions about the dynamics of prices in the future. In other words, price movements are random and unpredictable. This theory is set forth popularly in the book "The random nature of prices on the stock market"edited by Paul Cooper in 1964. Much has been written on the subject since then. The theory is based on the "market efficiency hypothesis," according to which prices fluctuate freely around real or intrinsic value. Another essential tenet of the theory is that Best market strategy - is "buy and hold" as a counterbalance to any attempt to "beat the market.
It would be ridiculous to deny that a small percentage of randomness or "interference" does occur in the market, but even more ridiculous is the claim that all price dynamics are inherently random. This is exactly the case when empirical observation and practical experience are better than high-minded theoretical research. These researches can prove what the theorist was going to prove, but they are powerless to disprove anything. It should be kept in mind that more often than not randomness is determined by the inability to establish systematic patterns or regularities in the dynamics of prices. The fact that many academic theorists have failed to discover these regularities does not prove that they do not exist.
All this scholarly debate about whether there are market trendsFor an ordinary analyst or trader they are of no interest at all, because they live in the real world, where they have to face market trends all the time. If the reader of this article is in doubt, look at any chart book (chosen at random) and the graphical representation of the trends will appear to you in all its glory. How exactly do the adherents of the theory of "random events" explain the enviable constancy of these trends if, in their opinion, prices are "serially independent", and what happened yesterday or last week has no influence what will happen today or tomorrow? And how do these scholars explain the fact that many trend-following systems actually make tangible profits?
So what trading strategy to choose?
How, then, can we recommend for use on commodity futures marketWhere the right timing is so important, a "buy and hold" strategy? Does it mean that long positions should be held during a bear market? And how does a trader know if the market is bullish or bearish? bearsif prices are completely unpredictable and there are no trends? What kind of bear market can we talk at all, if this concept itself implies the presence of a trend as such?
It is doubtful that statistical studies will ever be able to completely disprove or confirm at face value "random events. Nevertheless, technical analysis The idea of the randomness of the market is denied. If the market were really random, no forecasting would be possible. Strange as it may seem, the "market efficiency hypothesis" is closely related to one of the main postulates of technical analysis, which states that "the market takes into account everything". But theorists believe that if the market instantly takes into account any information, it is impossible to take advantage of this information. As we have already mentioned, the basis of the technical analysis is the fact that any important market information is taken into account by the market price long before it becomes known. Consequently, unwittingly, proponents of "random events" theory insist on the need to closely monitor price dynamics, but suggest that fundamental information should be neglected, at least in the short term.
After all, any process will seem random and unpredictable to someone who does not understand the rules by which the process occurs. Take an electrocardiogram. For any of us it is nothing more than the registration of some chaotic noises. For a specialist, however, every stroke on the tape is important, and not one of them is random. And so the dynamics futures markets may seem random only to those who have not studied market behavior. The mistaken impression of chaos goes away as one becomes more adept at analyzing charts.