laz asked me to post here. I am glad to do so. In any forum when people do not know each other nor have common experience and backgrounds, things go less efficiently than under other circumstances. Plainly, the efforts I make don't get across very effectively. For all my life I have given away my work and thoughts. As a person who has been judged by by phenomenal world contributors in many fields, I know that I have had a treasured life and it is coming to a close now. I have tried to be responsive to others in ET. On the other hand it is easy to see that here, it is difficult to carry on constructive mutual activities that can be very beneficial to all. Mutual effort is synergistic especially when all are considered peers based on their differing experience and ages. Market theory is one of several arenas in which I operate. I happen to have a flat curve relative to intellectual fields potential as measured by others who wished to make the effort. Markets can be used to make money. If you construct a dichotomous key to delve into which of mant paths to follow to become efficient in extracting money, you come to where this thread is operating as a major path to explore and understand. How it happens that there is a periodic (integral multiples of least contrast possible) relationship among market independant variables, is worth understanding. It is more worth your energies to go through duplicating the discoveries attendant to this. Here in this thread, it is made use of. For what ever reasons, I have always made discoveries in the fields in which I have been asked to work. The periodicity presence serves as a ranking mechanism of importance of the variable. The only conclusion possible regarding general market operation becomes one that only involves participants in the final analysis, specifically market behavior as a result of participant agreement. Decoupling A/D from market trends has been pointed out in this thread. And for fineness of observation it must be used. This is not a dilemma. To solve complex problems, a person must resort to elegance. I did for making money in markets. Binary numeration is required to minimize the differentiation (importance of differences) among unequally important variables. Time passes NOW by traveling from the future to the past. Money is only made in the present. So the scoring is set up to have "significant numbers" operating in the smallest base possible to minimize the contrast of significance. When all three variables "change" you are at the most significant place in the cycle. When two of the three variables change, you are at the next most significant place. Lastly, you are at the least significant point of change when only one variable changes. Change of trend direction is the most significant. 0 to 7 score and 4 to 3 score requires all three variables to change. The money velocity accelerates, maxes, and decelerates. The max money velocity occurs when two of the three variables change. This is 6 to 5 for long trends and 3 to 2 for short trends. treating price as a periodic function, this is the horizontal axis of the function. Thus scoring is perfectly symmetric in each and every way symmetry can be measured. All trades take place between people in toal dissagreement about everything but price. They agree on price. The minority of participants determine the trend direction at all times. This control changes most frequently and only half the time affects the trend direction (peak (4 to 3) and trough (0 to 7)). The other half of the time it affects the rate of change of money velocity. (ending acceleration and beginning deceleration at the axis crossover). I have heretofore only delt with beginner trading at ET except for introducing SCT, NLP, and PNI. Making money deals with optimizing money velocity of capital as it's first order of business. Scoring of investments in a portfolio leads to a singular process of switching out the lowest money velocity item and replacing it with the highest money velocity potential item. This is called cross over trading. There are seven equations (binary) that I originated for this purpose. They focus on scores 1 to 0 and 0 to 7 and 4 to 3. There is also available a mechanical C language program being passed forward that people use for this. Here are some caveats. laz is really doing a superb job. The application of this to any entity in any market is appropriate. By comparing and contrasting various fractals the optimum market operating point may be determined. If you take the time to understand, you will find that the market changes its operation point smoothly and continually. Since the fundamental periodicity of the market is determined by price, this alone is used to determine the pace of the market. "Taping" is the place to look to do the least activity to make money and the fractals faster than taping are where to look for the optimum money velocity. when you go past optimum, the reason for the breakdown is the preponderance of "noise". There are upper limits in markets for making money. This has not been discoverd as yet apparently. I know from experience that I cannot trade effectively more than 10% of the cummulative trading done in an equity on a given day. I also know that the float of an equity is one of the most important FA considerations in making money. Finally, macro market analysis does not work in any form whatsoever for making money nor optimizing making money. Analyzing ES futures index is not macro analysis. You can always use different fractals of one index as different items risk and money management wise. This is equal to doing FA on different equities to put them in your portfolio as independant investments. I thank laz for the work he is doing in this thread. I am commenting in a particular way here to just illustrate some stuff I have been working on since 1957. Learning to trade is not a book reading thing. Quite the opposite; it is a process. What I like best is the iterative refining based on 47 years of experience so far.
Bruce, Thanks for your work! That's really cool. I do wonder what it's like to have this as one's operational model (from a discretionary standpoint) while watching the market during the day. Still too complex for me. Is it DOM? Is it longer term (less resolution) so the orderly transition from 7-4 and 3-0? These things are not so clear-- I don't really have a def of acc/dis when I'm watching the market, except for a separate t&s window that tracks 30-car or greater trades. In my head it's the 'tables and placards' metaphor, except instead of placards everyone has water guns. You can imagine the look on the 1-lot water pistol guy's face when two 400-lot supersoaker 9000 watercannon shots get loosed against him. I'm glad everyone enjoyed the exercise. --laz
Laziz, I've looked at the histograms again...statistically, looking at the means and standard devs. you only have four real patterns there, not eight. I think this whole deal can be simplified. Dealing with three variables and looking to two of them switching at any given time gives rise to the problem we have right now, 8 Rules at the most basic level with as many as 8 possible outcomes per Rule. Although we can use frequency to help guide our decisions, the more possibilities the greater the uncertainty. We want to reduce the uncertainty as much as possible. The way to do that is to use the Rule base for one thing: determine how Price and Acc/Dist are related. Volume is separate because, as DBphoenix has noted, Price can move with or without Volume. But, reading any common definition of Acc/Dist reveals that it is used as a confirming indicator to Price. That is why I think you should link up those two and treat Volume separately. Certainly, it is nice to see the 4-pd. sum spike at what DB terms 'climaxes' of activity. But that is separate from the story Price and Acc/Dist are trying to tell us. I'll give it a rest for now.
Yeah, I think that's what I was getting at with the commentary in the doc about not having one of the items phased correctly. Jack gives the clue above about finding the correct fractal in which the cycle is operating (i.e. 'taping'), which is something I can eyeball but have no way of programatically determining at the moment. In the doc I thought it might have been a/d, but the operating fractal commentary makes it clear that price is the issue. The more I figure out, the more I figure out how little I really know. --laz
That shows to detractors of Jack that he can bring some value . Now I can tell from my equations that considering pv is not enough to achieve a universal modal stock market: local and descriptive approach (see also http://www.elitetrader.com/vb/showthread.php?s=&threadid=29554) whereas the market thinks globally and only act locally: if you only think and model locally you will never get the big picture. But I won't tell more because I want to keep absolute secret about the details. Nevertheless I may show a more simplistic model with the advantage that it is more comprehensive than my above sophisticated model but I don't remember where I put the files after crashing my disk so I will do so when I will find it again .