The BELIEF in Technical Analysis is a self-fulfilling prophecy. If you get enough people convinced that it works in spite of the evidence, there lies the truth. Without it, the market wouldn't care at all.
The science of Technical Analysis (TA) involves the normal five step process of the Scientific Method (SM) To hypothesize about TA takes a mind that is able to use the SM properly. This means that probability is not involved because probability is not part of the SM. The hypothesis made is usually modified repeatedly as the scientist narrows the chase to the theorem or theory he is seeking. I liked the way the SM resolved the aspects of my use of TA in trading. I focussed primarily on the null hypothesis by necessity and I made sure I handled the full spectrum under consideration. This, for me meant disproving the null hypothesis for all cases. By moving from probabilistic mathematics out of necessity, I found I had a real and comprehensive methodology that did not contain the usual traps that are reported out by most persons who are examining aspects of TA without the benefits of the SM. Not everyone is trained in SM. Consider, for example, a person with an MS in CS. There are two S's and still that does not mean that the person has had any time spent studying the application of the SM in or out of the fields of science to which the SM normally applies. Broadening the application of the scientific method has been done and it is the logical this to do whenever possible. It is not, however to be done haphazardly or just pragmatically in a superficial manner. Saying "TA doesn't work " we all know is not a hypothesis nor is it the null hypothesis for the consideration of TA. It is easy to test this statement, presumably by finding an exception but usually the exception is just a time bomb waiting to go off so to speak, scientifically speaking. It is not a probabilistic issue when the chips are down and in the final analysis. This is just a superficial kind of approach that would prove nothing with respect to a theory or theorem set. An earnest effort to become informed about TA is important for anyone since its application may be possible for them in trading. Applying TA to trading is not a possibility for many people as they have told us and told us frequently. For those where it is not possible to use it, their views are very important since they serve as examples of what can happen to anyone and for the variety of reasons they state. This failure list is a list of the traps a person must avoid when applying the SM or any reasoning to the application of TA. Getting trapped by personal mental processes is very possible as we see by looking around us and in our personal journals of our work in process. I surveyed TA to find the sources of TA where the created TA as a theorem or theory was derived in a manner that involved the SM. If felt that this was logical and orderly and that it might be possible to build a heirarchy or body of information that was rational and logic. I was amply rewarded. I also used my training to hypothesize and conduct SM experiments. Experimentation is not usually conclusive but it has a powerful value to the person doing the work. It is similar to problem solving (PS). PS is common because of its pragmatic reach and applications. Most problems are not solved easily. They would not be problems were it ieasy to solve them. Most people never do have the opportunity tosolve problems in the course of their work. Those that do PS often only achieve partial solutions, or to put it another way, they change the original problem from one thing to another that does lend itself to further work. Applying science to TA by the use of the SM, for me, led to a series of steps where the hypothesis gradually changed to give me a focus on the kernel of the opportunity that TA represents. As I resolved each small aspect that appeared to be processable, these aspects helped form the body of elements that I now have that are integrated. Dealing with TA scientifically is one big, complex and immense job. the people who worked to create TA where largely inspired by two themes: they wanted useful tools or they wanted a product or service that had market value. I regard Welles Wilder as in the pormer group and a person like Oliver Velez as being slanted more towards the latter group. Modelling is not usually scientific it turns out It is more oriented to the application of scientific principles (not theory or theorems) to pragmatic demands. Securitization of debt instruments is an axample that is unworkable scientifically as we all see. the bomb is now going off as we know. Development follows modelling and it is even more remote from science. It is computer "science" a bucket that was just dropped into the academic production line when computers were invented. It is a plumbing degree for hardware plumbers to make software for the hardware to operate. Coding has no relationship to the SM. Coding makes hardware work. Maybe this thread will deal with creating TA using the SM. That would be a very profitable kind of work for either of the two themes. It is one heck of a trip to go through the disciplined effort to use the SM to come to a body of TA that is comprehensive and has no time bomb associated with it. The forks in the road turn out to be the less travelled ones.
This looks like a brief retail oriented list. These are probably sales oriented gizmos that are designed to sell well. Why don't you consider posting a TA design oriented list and a list that TA strategy application oriented list.
You've hit the nail on the head by saying TA/SM has a relation to PS. Most people, let alone traders, are unfamiliar with the processes that involve in depth PS. The main problem with the average trader is that they do not want to spend the time doing the necessary grunt work on their own to prove OR DISPROVE a program works. If it takes longer than an hour that is too much time. They would rather just blow off any reference to the possibility and continue to trudge along in the darkness. I suppose they feel it's safer that way. There are threads that outline the SM. Here is one: http://www.elitetrader.com/vb/showthread.php?s=&threadid=80582
What do you think is the criteria for TA systems? Why is none of the above software suitable for it? Has anyone read "Does The RSI Give you an Edge" by Larry Conners and Ashton Dorkins in the November issue of Technical Analysis. Any comments on that article. I have played with a 3 day RSI before, and had some luck with it but never with a 2 as that seems too short of a time frame to me.
You have it backwards. TA has been exhaustively tested by many, and it is constantly found wanting. The onus lies on the TA believers to PROVE it works. Null hypothese and other things have no value. Either a multiyear application of an indicator across multiple values across multiple instruments shows a SIGNIFICANT outperform or not. This is statistics and fact. Quoting someone from Tech An of S&C has little meaning. They are known to cherry pick examples. And many of these people are trying to get people to thier websites, so they cannot be considered as objective examples.