http://www.elitetrader.com/vb/showthread.php?s=&threadid=134037&perpage=6&pagenumber=444 TW,if you understand all the macro stuff,you will like this thread,ii read it but most of it's over my head
It's not over your head you just haven't applied yourself in that area yet. In the next three years commit yourself to only making 4000 posts and the rest of the time getting your deep study on. So while we're discussing it lets give that kind of information a broad stroke review... it's pretty much worthless. There are so many "opinions" about what this economic indicator means or what that meant; it's no different that T/A... ideas and opinions without basis. So what's the point I'm slowly trying to drive home here? Traders need fewer belief biases i.e. "I think the dollar is going to tank because we're likely headed for a double dip recession..." or "Based on RSI the EUR/USD is 'over-bought'." So what's the answer, what's the alternative? I can't tell you, because my trade model is played out. But I can tell about the process I took which should be valuable. When I was trading professionally I took an approach that was growing in popularity and becoming more accessible, but I structured it in a completely different way. When I started my research I did it in every applicable field but Finance. I didn't want to bias myself with a particular approach. What I knew for certain was, if there was significant research existing in the financial arena for this application it would already be so excessively implemented that there would be no competitive edge. This was proven over time by the lengths the big houses were using to reduce execution latency. Once I understood how the application could be used I set out to build several models and approaches. I ended up with about five that appeared to have merit based on significant back-testing. At that point I dug into the financial research. I looked at every possible scenario for implementation, some of my ideas were already being used, there was significant research that both supported and criticized the approach. Those models were kicked to the curb and I focused on the ones that "worked" but had a lower correlation of trades to the others. Effectively I ended up with an edge. Or, a "small" edge... I'm not dumb enough to think I was the only guy trading this model, but it was something that wasn't well represented at the time. Today... well that is a different story. The model and approach have been beat to death and returns are dwindling and correlations are increasing. This is the nature of the beast. To succeed as a trader my belief bias is this (and I do have support for it...) You need deep knowledge: Understand how the indicators (TA & FA) are derived. When you can do the math in your head or on a scratch pad, you'll begin to see how indicators are going to shape up (especially in TA) before it happens. Be able to assess in your head how blocks of activities and events effect price action, forget about single events. You need to see the markets as ten different chess boards that ultimately affect the play of a single game. See "chunks" of indicators and data, not individual numbers. Before you can really attack knowledge acquisition you have to bring your skill set up. Practice anything that applies; for most of us that means getting back our basic math skills and/or acquiring new ones. When you have skill, and knowledge you can build... In this case you're going to be building models and approaches that you hope bring you an edge, but there are no guarantees. To help assess the quality of a concept you have to research the hell out of the thing, both for supporting and opposing documentation. Both will help you determine if there is a huge fundamental or usage red-flag and hopefully if your thinking is unique enough to have merit. One of the biggest mistakes new/retails traders make is thinking if "this guy trades this way it must work...." Nope. If it did work at one point, it probably doesn't any more. You can debate efficient markets until the cows come home, but one thing we do know is when an approach is widely implemented its returns diminish. Now that doesn't mean it wont cycle back around into favor, but not knowing when and in what form leaves you at a disadvantage. I.e. Trend trading parameters that were useful in the last ten years may not work in the next ten and we won't know what parameters were useful until we can LOOK BACKWARDS and check it. Not very helpful when you have to "trade it forward".
I'm glad you've finally got around to "What's really on your mind". This is a "retail" board. We all think we're part of "the fat tail". Some of the posters in this thread have paid their bills for years based on what they do in markets every day. No one is going to write a White Paper about it. As you have said, when you find an edge it's best not to disclose it. That doesn't mean good advice can't be given. I read this board for ideas and nothing more. You paint a great picture of a well thought approach. Your comments on what happens to widely implemented approaches resonate. As does the need to research and really think clearly and creatively about the bigger picture. I posted the quip about Kase because her firm is now posting outlook predictions for the world to see. Yes - It's being done to hawk her studies to subscribers. eSignal (bottom of the barrel / lowest common denominator) is now offering them. I'm not stumping for her or her book. I suspect she went retail in 96/97 because her institutional customers were becoming less and less impressed with her tool set. The most recent forecasts look based on the same methods used in her book. Much of what is behind their approach (excluding the better statistical math in their indicators) has been over sold to retail for years and also widely panned in many places on ET as well. My instincts are that these predictions are going to come up way short which will further lend credence to your point of view. Its worth following. Back testing / forward testing make no allowance for changes in market dynamics. This raises questions about any model shaped based on testing static concepts. The problem on the pro side of the ledger if you can't model it, you can't sell it to institutional clients. Big money only bets on "sure" things. So far the brightest and best haven't been able to build any "Market Weather Forecasting" tools that haven't blown up due to assumption failures related to shifts in underlying participant psychology expressed as market dynamics. I've never been a "system" person, but I've always tried to court the wide view you reference here. I believe technical's have to be used to point the way overlaid against economic fundamentals / geopolitics / participant psychology. Chaos is the fundamental characteristic of markets. History has shown that many times over with all the burst market bubbles. Some of the advice posted in this thread (which you've panned) is based on that understanding with out verbalizing it. Like everyone else what I write here is merely my opinion - You know what they say - Everyone's got one I can't defend it with the rigor you want, but that doesn't mean it isn't valid. Thanks for your honesty and for the ideas you've put out here. Anyone who works hard as something can't but help develop strong feelings and convictions. This thread is shaping up to be a nice counter balance to what is fed to noobs that trading is easy and simple. On this I point, I think we can all feel we've added value here.
Yes and no. Lets presume you are going to take the offer of an instrument. Any instrument would have about 7. Pick 3 adjact ones for your trading. For ES I use a barframe to display fractals. I can "see" 4 or so fractals on a barframe. A pane is used for this and it is a custom frame stored on one the plaforms I use. I add more panes that reralet to this pane. As I annotate on the main pane, the annotations are carried over on to the associated subordinant frames. There are other panes that have the same data expressed in other ways than P and V. I follow the four games played on the DOM so I have the dom up as well. Part of the monitor and analysis of the games on the DOM involves the T&S pane. I also connect this order oriented stuff to the smart money's snetiment. There I examine the relationship of smart money on cash and smart money on indiexs. I also look at the OTR and itsconnetion to the T&S. So I have a p,v pane for OTR. The above block of stuff is all leading price just as volume leads price. Each thing mentioned has an order of events. The other half of my display deals wirh the parallel relation of leading lagging indexs. I also have, with panes a faster leading index of the ES. Then there is my second system which is operating. It has a core and three shells. The above is a core and shells also. The "Sweeps" Chart" composed of nime tables and their indicator signals named in the cells of the tables, gives you the firing order of about 70 items all of which have their order of events and whereby these streams are interelated and interlocked in terms of orders of events. Simply stated it is like what you look at, when you lok , etc,., to drive a car or fly a plane or sail a racing yacht. My most complex car has to banks of cylmders, 12 matched fuel injectors. I, electronically select, which of 3 tranmissions I wish to use in given traffic. The saftety systems are all dual as they are on aircraft. I need to know if my rated tire (rated for high speeds) pressure is changing to any extent. Dual fuels lines and pumps are considerations. I also can turn off one bank of cylinders if need be. Burning rubber to accelerate when in the 160KM range has to be guarded againt if I am using the "sports" tranmission to pass a redneck driving a picup. I need to keep track of my intrument panel at all times. My four suspensions automatically adjust to assure stability when forces are present. If someone changes where they are sitting, an adjustment is made. I have one master hydraulic pumping system which feeds the other pumpimg and hydraulic systems described above ; all portions are dual as I suggested. I am told how the various parts are working. My master mechanic EXPECTS me to tell him what I saw and what I heard if and when he is services my car. My car talks to me as well. when I change batteries I have to go through resetting the lauguage of the system defaults to a german male voice. If I do anything wrong it is explained to me. To have a display for trading, you pay for it. To have a car you can drive effortlessly, you pay for it. To have an expert system that continues to take the market's offer fully you have to have a display, a differentiated mind, always be on the correct side of the market and not exceed the capacity of the market by more than five times the market. A potential trader steps into the picture and stands at point A. He is going to point N at which point he is an expert whose mind is fully differentialtes and he can take a day's profit to a car dealer and do the specs for a car that then will be built and shiiped to him for his test trials. I know several guys who have waited over four months in delays for delivery. A also know how each went from point A to point N. Lets say you make up a log that contains the columns of the elements in the Sweeps Chart. I DO. What I fill in as I sit at point N is 10 to 12 pages of logs per day. I email the logs to others for thier bebriefing. I also attach my annotated charts andI attach my personal written debriefing. Now I will start at point A and I will include multipage bibliographies for each structural comment I make in terms of both Behavioral Finance and Technical analysis. It is simply a cut and paste fdor me whereby I have fingertip control of the INFORMATION. First I will explain how science works. You all gret to see my science results in three ways every day of your life regardless of your lifestyle. Adding trading to the mix, youwill get to see four problems solved using only one scientific approach. For one of these our team was awarded new science program of the year by the NSF. That year winners of the NSF science prizes got a prize that included working with our team for three months before they went to their tier one science oriented universities.
This thread is oriented to the mind. Traders partner with markets and they do their job. This requires a fully differentiated mind that supplied the infernence, on demand, when the market displays something to be sensed. BF calls these items EVENTS. You go here: http://www.behaviouralfinance.net/ And you click on BF or BS (in blue) located to the left in the main pane for learning about BF. In that citation you find: "BF or BS? If behavioural finance is to replace the efficient markets hypothesis as the most widely accepted paradigm, it is not sufficient to simply find flaws with the EMH. Rather than representing a unified theory, behavioural finance often stands accused of consisting of little more than data mining for anomalies followed by the search for a behavioural explanation. Fama (1998) concluded that "[m]arket efficiency survives the challenge from the literature on long-term return anomalies. Consistent with the market efficiency hypothesis that the anomalies are chance results, apparent overreaction to information is about as common as underreaction, and post-event continuation of pre-event abnormal returns is about as frequent as post-event reversal. Most important, consistent with the market efficiency prediction that apparent anomalies can be due to methodology, most long-term return anomalies tend to disappear with reasonable changes in technique." Kahneman and Tversky have shown empirically that people are irrational in a consistent and correlated manner. However, the case for the EMH can be made even in situations where the trading strategies of investors are correlated. So long as there are some smart investors and arbitrage opportunities, they will exploit any mispricing and the irrational investors will lose money and eventually disappear from the market. Martin Sewell" Take a hi liter and hi lite the BF A, B, C, and D. Use a ballpoint to circle and name the A, B, C, D. Print, punch and put in a three ring binder. you have been told to journal your emotional flags. You qwrite down the screw up your approach is showing when you heve such emotions. BF says to you on this post to make a reasonable change in your approach to remove the flaw in your approach that caused the emotional upwelling. You are at point A standing there flat footed. Get a second hi liter and underline the words: "continuation" and "reversal" You will find the only thing noted in your journa; is screwups related to either continuation of reversal. the trader deals with events from the market. All are precisely realted to continuation or reversal. _______________________________ Science is a beautiful thing. Markets are systems that are operating scientifically. The description of the markets is simply a data flow describing the acitivites of the participants. The trader divides this flow into a series of events. As you hi lited events have "pre" and "post" behavioral measurables. The market is measurable and the trader is measurable. The science of the market comes from a scientific statement that is kindred to the scientific method. Google Keynes and couple your google to "like kind". You are looking for the foundation of scientific statements and you are going to GET the scientific statement that wholly defines the market. I had to read 7 pages on day 1 of my expereinces that began in 1957. What I read was sort of casual so I had to do what Keynes suggested in Paradigm Theory. From the full statement, you go to work in only a deductive manner to find the hierarchy of principles and their corrollaries that infill among the principles so that you have a fully differentiated system that addresses every possible event in market operation. In real time, doing this is a very brief period. Pragmatically, it allowed me to set up an account and contribute half my salary to the account and be profitable immediately. Toys, travel, marriage and quitting my job soon followed. As you see there are questions on the table from others and still others answered those questions. One Q from Nodoji was a terrific one. She gifts a trade to another, the person gets 20% of what was suggested. Why didn't it work out well? The person who couldn't hold to get profits could do this: 1. copy the BF quote, etc. 2. add a sheet on the trade to the 3 ring binder and debrief the trade. 3. write down the A underreaction. 4. write down the B pre event (meeting target) abnormal behavior causes. 5. Write down the irrational behavior of BF D. Now do the reasonable changes to the technique. 6. Say back the trade to nodoji. 7. Say to nodoji each step of the treade, especially the words: "NO EVENT". 8. Pass the 20% point in DEBRIEFING with a debrief of what wasn't That (the non event) that caused to exit early. 9. Discuss this "invention" that was NOT part of the trade. 10. discuss making a log of the trade IN ADVANCE. 11. do a plan to "check off" the parts of the trade as they are observed. 12. Repeat 20 times for next 20 trades. Now the three ring binder is getting thicker. Neither trader is in the market much of the time. So debrief on why neither are in the market and what market offers were missed as a consequence. Maybe even turn to thinking about how the market works.
How does a scientist step off from point A to go toward point B. He does an "in kind" hypothesis set according to Keynes. H1 and H2 result. They are look alikes. They have to be to be in kind. Dood/Granville spoke in the terms of the markets variables. They were part of the 30 some people who did the same. I noted the people, dates and What hierarchical item, Why they did it, basis of item, How item works, and the date pf publication. Here are these sheets with the development of trend monitoring and analysis noted. http://www.elitetrader.com/vb/attachment.php?s=&postid=2960978 http://www.elitetrader.com/vb/attachment.php?s=&postid=2960982 http://www.elitetrader.com/vb/attachment.php?s=&postid=2960992 For me the red dots hi lite the items I use and they are numbered so I could put them in an Excel hierarchy. Attached is a page of references for the above timeline charts. Science dictates deduction and not the use of induction. Induction usually involves data processing in the statistical sense. You can use the timeline sheets to trace how the statisitical analysis originated for trend monitoring and analysis.
The form of H1 and H2 are identical. four words that are unique are used: Increasing, decreasing, continuing and changing. In English, these are the same parts of speech This means Keynes' insistent on "in kind' has been met. By looking at markets from top to bottom and drilling down to the smallest parts, a complete hierarchy of pieces can be selected. The pieces get grouped and more principled items result. The groups form bigger pieces and these can get large enough to form the principle pieces. The form of the H! and H2 is an IF...., Then..... statement. Thus we have four market measures all ending in "ing" and these measures are plugged into an IF...., Then format. The consequence of this effort ditates the type of mathematics that must be used to build the system. So far I have never seen a system require more than 5 levels in the hierarchy in the structure. Data is sent through the structure. OODA has a structure and I probaly should post it and its detailed feedback loops. This OODA routine takes laps as was seen. In the process of lapping, statisitics results DO arise and the OODA prodess is typified as an inductive process. A lot of quant type work under OODA auspices has lead to some really strong viewpoints on distributions of events and, in particular, extreme events. The result of this statisticulating is very significant to the OODA users in the CW context. O6MAY10 was evaluated in a lot of places and the one I like best is filled with wrong observations BUT the coloration of the segments of the explanation is an identity with the pattern produced by H1 and H2 and its descending hierarchy. I have used it often as an illustration. H1: If Volume is increasing, Then the price trend is continuing H2: If Volume is decreasing, Then the price trend is changing For trading this is a hold/reversal strategy. A trader uses either H1 or H2 at any given activity of the market. Science uses the Quenn of tools, mathematics, for its operation. Dow Theory arrived and it was quickly followed by the two men who created the basis of creating H1 and H2. If you are a student in computer science or a student majoring in maths, you immediately recognize that an Algebra application is at hand. The mathematics was invewnted after Rothschild and before Patten on the timeline. Coincidentally, the same mathematics is used for gaming amd for computers. We have gone from nowhere (point A) to a place of beginning (call it point B on the way to point N). At point B we have our paradigm Hypothesis Set (HS) and the paradigm measures (PM's). It is a nodes and links system. Events are at nodes and links ar used to pass time between each node. We must use Boolean Algebra and the measures are velocity (in a vector form) measures of variables where one variable leads (V) the other variable (P). BF that you have hi lited and circled proves how the human response shapes up in terms of human signals and how reasonable change is put in place to improve human performance.