Here is a search result from Google.com for the author of the paper. http://www.lbrgroup.com/ Q We are extremely proud of our friend Gary Anderson, this year's recipient of the Charles H. Dow Award. Gary follows a classic Wyckoff methodology, but has put together his own creative relative strength observations based on over 30 years of experience in the markets. Click here http://www.equitypm.com/TheJanusFactor.htm to read his award winning paper. UQ
Lawarence Harris: The Winners and Losers of the Zero Sum Game - The Origins of Trading Profits, Price Efficiency and Market Liquidity http://www.torontometastockusers.com/zerosum1.pdf . Q In the process, we shall obtain a more complete understanding of the origins of market efficiency and liquidity. UQ
Apologies for the delayed response. If we stick with our premise that TA is evolving through a feedback relationship with the market and that it is evolving into something so complex that computers will be required to deal with it, we have no guarantee that market efficiency would improve. First, said computers would be programmed to execute in accordance with software construed on the assumption that TA is valid. This assumption may be erroneous and given computers' lack of ability to alter level of perspective (to assume an overview) the result would be chaos. On the other hand, perhaps TA is valid and will evolve by way of said feedback effect; this would almost certainly result in exponentially increasing evolution that would have unforeseeable results - almost certainly chaotic again. TA wouldn't become so complicated that it would cease to exist until it reached the point of computers' inability to digest data. But that point would be reached sooner or later in an exponentially developing complexity situation. At this point, TA might have become so bizarre as to be unrecognizable as such. A return to fundamentals wouldn't be easy to effect. The computer-driven system would first have to collapse - and that would probably be catastrophic. Also, consider this: If TA is valid, there must be a perfect TA paradigm. Said paradigm would be universally applied and there would be no reason for the existence of exchanges; if someone had to sell a holding the agent institution itself would buy it if it were a TA 'buy'; if it weren't a TA 'buy' , nobody would buy it. Exchanges would be irrelevant; everything would be done in-house. On the other hand, even in feedback systems, points of equilibrium can be reached - and that might be your point of absolute efficiency. These are just some possibilities that come to mind. Just idle speculations of course.
My 2c Since it's origins, T/A, like any popular concept, has changed (evolved if you like). The major elements in it's change are - increased choice of methods and analytical approaches - increased importance of feedback loops - TA affecting the price which in turn is affecting TA which then... etc - as a by product, the price manipulation could also be included here It has been argued that this will lead to increased complexity with various, not very positive, outcomes for TA and it's users. I put it to you, that in essence, TA has not changed a bit. - While we may have been forced to use different or new methods the complexity plays a role only in the selection process of the right method to use. Once we select such a method, our TA is as simple as it always was. - The feedback loops are an important element of TA but not necessarily the reason for increased complexity. It may appear that this creates one or more additional variables in the model but I believe the impact on complexity will not be great. Suppose we have a model of a process with, say 10 elements that govern it's workings. Introducing a new variable into such model will not greatly affect the complexity of our model because the greater the weight/impact of new variable the less significant other variables become and our model can then be simplified by removing some of the old, less significant or obsolete variables without affecting statistical results of such model. In extreme cases this may actually lead to the model becoming LESS complex. For example, if feedback loop becomes a major force behind the price movement, then it's patterns could become more predictable than they would be without the feedback loop. - the fundamentals still remain as the regulating force for all TA as we've seen during the tech bubble and on a smaller scale everyday in the market. It all comes down to a chart, patterns and supply and demand. The patterns may have changed, but can they be much more complicated? Our interpretation of supply and demand may also need to be changed to include the new driving elements but this does not mean complexity, it only means the difference. The concepts become complex only if we want to adapt to the new changed environment while still hanging on to all principles held in the past - that is we'd rather enrich our old theories by adding new concepts than replace them with new ones hence the burden of complexity. But I suspect that the new TA model is as simple as the old one and the difficulties in finding one are caused by the same problem which we had to overcome when adapting the first model - much of the knowledge must be questioned and "unlearned" before we can adapt the new concept.