This is the answer to my question: 1) Confirmation Bias The confirmation bias is a tendency to seek information to prove, rather than disprove our theories. The problem arises because often, one piece of false evidence can completely invalidate the otherwise supporting factors. Consider a study conducted by Peter Cathcart Wason. In the study, Wason showed participants a triplet of numbers (2, 4, 6) and asked them to guess the rule for which the pattern followed. From that, participants could offer test triplets to see if their rule held. From this starting point, most participants picked specific rules such as âgoes up by 2â or â1x, 2x, 3x.â By only guessing triplets that fit their rule, they didnât realize the actual rule was âany three ascending numbers.â A simple test triplet of â3, 15, 317â would have invalidated their theories. 2) Hindsight Bias Known more commonly under âhindsight is 20/20â this bias causes people to see past results as appearing more probable than they did initially. This was demonstrated in a study by Paul Lazarsfeld in which he gave participants statements that seemed like common sense. In reality, the opposite of the statements was true. 3) Clustering Illusion This is the tendency to see patterns where none actually exist. A study conducted by Thomas Gilovich, showed people were easily misled to think patterns existed in random sequences. Although this may be a necessary by product of our ability to detect patterns, it can create problems. The clustering illusion can result in superstitions and falling for pseudoscience when patterns seem to emerge from entirely random events. 4) Recency Effect The recency effect is the tendency to give more weight to recent data. Studies have shown participants can more easily remember information at the end of a list than from the middle. The existence of this bias makes it important to gather enough long-term data, so daily upâs and downâs donât lead to bad decisions. 5) Anchoring Bias Anchoring is a well-known problem with negotiations. The first person to state a number will usually force the other person to give a new number based on the first. Anchoring happens even when the number is completely random. In one study, participants spun a wheel that either pointed to 15 or 65. They were then asked the number of countries in Africa that belonged to the UN. Even though the number was arbitrary, answers tended to cluster around either 15 or 65. 6) Overconfidence Effect And you were worried about having too little confidence? Studies have shown that people tend to grossly overestimate their abilities and characteristics from where they should. More than 80% of drivers place themselves in the top 30%. One study asked participants to answer a difficult question with a range of values to which they were 95% certain the actual answer lay. Despite the fact there was no penalty for extreme uncertainty, less than half of the answers lay within the original margin. 7) Fundamental Attribution Error Mistaking personality and character traits for differences caused by situations. A classic study demonstrating this had participants rate speakers who were speaking for or against Fidel Castro. Even if the participants were told the position of the speaker was determined by a coin toss, they rated the attitudes of the speaker as being closer to the side they were forced to speak on. Studies have shown that it is difficult to out-think these cognitive biases. Even when participants in different studies were warned about bias beforehand, this had little impact on their ability to see past them. What an understanding of biases can do is allow you to design decision making methods and procedures so that biases can be circumvented. Researchers use double-blind studies to prevent bias from contaminating results. Making adjustments to your decision making, problem solving and learning patterns you can try to reduce their effects.
Notice that the closing price tends to end up far away from the open price. Thus many people intuitively believe that there must be some deterministic forces at work that moves prices and causes them to trend away from the opening price. This intuition is wrong. I think a common intuitive misconception is that if the market were random then prices would oscillate back and forth and pretty much end up unchanged. So why do prices tend to close far away from the open in the market everyday and in random series of data as well? From wiki: Intuitive metaphor for Brownian motion Consider a large balloon of 10 meters in diameter. Imagine this large balloon in a football stadium or any widely crowded area. The balloon is so large that it lies on top of many members of the crowd. Because they are excited, these fans hit the balloon at different times and in different directions with the motions being completely random. In the end, the balloon is pushed in random directions, so it should not move on average. Consider now the force exerted at a certain time. We might have 20 supporters pushing right, and 21 other supporters pushing left, where each supporter is exerting equivalent amounts of force. In this case, the forces exerted from the left side and the right side are imbalanced in favor of the left side; the balloon will move slightly to the left. This imbalance exists at all times, and it causes random motion. If we look at this situation from above, so that we cannot see the supporters, we see the large balloon as a small object animated by erratic movement.
So I ask this question again: Assuming a mostly random market, how do you make money? For me: (1) I don't have the patience for buy and hold and I cannot bear the regular 30-50% drawdowns this strategy produces. (2) I don't have the quant/programing/math background to do market neutral relative value arbitrage. So I am left with (3) trading off of undiscounted news/earnings. The past two days for this strategy were a great example. ES popped 30 points in 20 minutes when CNBC said ABK's AAA rating would hold on Friday. And again today we saw huge rallies in MBI, ABK, and index futures when S&P reaffirm the AAA ratings of MBI and ABK. Those clearly were not random moves - but you have to have the patience to wait for that kind of opportunity.
Hmmm... last I remember seeing you surf, you were avoiding the thread where you asked for a concrete example of TA working and I offered one. Don't bother denying it. I'm not going to do the searches necessary to find it. It was the last long discussion about TA. It is true - TA bashers want to convince themselves that human group behaviour isn't predictable, despite overwhelming evidence to the contrary.
The thing about surf that I don't understand is that he constantly says there's no such thing as trends (patterns) - he then uses that as a justification for countertrend and mean reversion trading. If there is no such thing as a trend then why would you countertrend trade a trend that doesn't exist? And if markets are overwhelmingly random then mean reversion is a meaningless strategy - is it not? If the mean is a random number and deviations from the mean are random, than what 'edge' can mean reversion strategies offer? Mean reversion strategies have caused more blow ups than any other. LTCM, Amaranth, VN, and ET's own Schindler are popular recent examples.
I think that one very important and salient point you are missing is that although markets have an unpredictability component in the short run (variation about the mean), the mean itself is not zero. Therefore, the long term bias is not random, rather the central tendency favors those who buy and hold long term (similar to a dc offset in electronic systems). If it were zero, then there would be no hope for anyone, including buy and hold over the long run. The fact that the mean has a slightly positive offset over the long run is what allows those who buy and hold to benefit over the long run. This in itself is a mathematical edge that anyone can benefit from with much less stress (less risk, albeit lower reward). The more diversified the portfolio, the more valid the mean offset becomes (i.e. indexes vs individual stocks, non-correlated assets vs. correlated). It's also one reason why people who look at long term charts can not fathom the chart to be one large random phenomena (according to their expectations of what random should look like). The long term charts tendency to drift upward can be expected, even using random models, if the entire model is correctly applied (including a non-zero mean).
For making money Taleb is humor. As Prof says, reading and selling his stuff is the best idea for carrying on conversations. It may be a good idea to give another person's comments some credit based on their aurgument (meaning developing their views). We understand that you are "printed" vis a vis your views. A lot of people posting have read what you are suggesting and their viewpoint stands as you read it. Taleb is "popular". A lot of financial authors are writing from the viewpoint of observers or those who have left the financial arena as practitioners. Taleb makes the point that using induction to build a strategy is pointless. On the other hand what does he say about how various types of orders affect the markets. Lets look again at the MIT BU data (2003). Or look at the frequncy of occurances of events Atchuler's scripts in wealthlab. The noted events in Taleb, MIT/BU and Atchuler are all rare in sigma terms of distributions. Honing in on making money and getting results (meaning a strategy) requires operating where there is money to be made, namely everyday all day long. You point out that I "wrote a book" when I reviewed your commentary and posted it as an attachment in Word. I didn't actually. What I gleaned was a key ingredient of how markets operate. They move from one operating point to another all the while offering money to traders. Generators that use several variables can duplicate the market and trading it to extract what is offered. Were the MIT/BU group to have done more with the real data they could have learned the relationships of the variables and how to work with the market variable feed as non stationary. Taleb couldn't. Nor does Atchuler. What does it take for experts in analysis to be able to turn out results that are relate to making money continually and to the extent that it is offered? You have worked at it for a period. Why has it not printed on any of those mentioned of you? It can be seen by the comments of many here what is understood and what is not understood. Knowledge is acquired by understanding what is important and what is going on. We all take steps to learn and do better. Learning to note and digest the MIT/BU was, for me, writing the math on the back of one of last year's Larson calendar days. I paper clipped it to the printout of the 2003 MIT press release. 2 cubed, 2 squared, and zipf's 2 to the first power all used in power series. About five or six terms are what is needed to get to the sigmas where frequent occurances of events are at play. This all dictates how to trade. A person, for black swans, has to be on the right side of the market. A person , for less rare events, has to be on the right side of the market. A person trading when the sigma part of the curve is near the middle, has to be on the right side of the makret for that money velocity offering as well. You state a pair of beliefs (your beliefs as a trader) as an "OR" and they happen to overlap. In any event, they can be used all the time under any event to absolutely protect the trader at all times. But only if they are measured in the present and used as what the market is telling you instead of what you believe. This and this alone is what takes a trader from trading with entries and exits to trading using the pair of statements in a market context. Look at big money and its performance. Big money IS in the market all of the time BUT big money is only on the right side (NET of times) enough to only make little of what the market offers based on the pair of statements that define, to a Tee, trend trading. you trade like big money except you are on the sidelines a lot of the time and are exiting and entring based on other than the pair of statements. Running a very tight ship is required when traders do trend trading. Big money does not trend trade as we all know. Big money trades based on induction instead. Taleb deals with their screw ups on rare occasion. MIT/BU looks at big money to discover what naturalists (physicists in this case who are coaching economists) can do with a huge data set covering four years and "all" trades. They know what # of shares cause what price shifts once in a while; they know about the number of days the number of shares will be happening; they know the distribution of big money by big dollars. Nature prevails and it is considered a secret revealed. The nitty gritty in the trenches for any trader of any size is what a trader needs printed in his mind. It is just like a spot on your bathrrom floor but it is important for extracting money vis a vis trend. Taleb, MIT/BU and Atchuler did not build minds based on trend trading ;either theirs or yours or others. Trend (price movement) works like a well oiled precision machine. It operates efficiently and effectively all of the time to express what the market offers to traders. Market pace versus volatility shows normal distributions. Since it does, then there is reason to know that the market migrates from operating point to operating point and, further, it does not jump around. Non stationary data flows and each slice of it (complete sets of market variables) drill on down through at least seven pragmatic fractal levels. This is markets operating and they are tied to bedrock and NOT determined by induction. You see them as floating islands of risk and they could be Tsunami prone on rare occasions. you use data to find "tells" for advantage during unbalance. Astronomers, during the super nova (seven days) in "87, did not believe their data until they ran it enough times to overcome their "prints". Simply stated the Universe was accelerating its expansion and not contracting as they thought. It is fun to stand in that place (physically) and here a participant describe it. What I am saying about the structure of markets is that they trend and, if you use only certainty based mathematics, you can measure trend and trend overlap and, further use the other half of the two statements to reverse when appropriate in order to continue to extract profits as each (non opposite) statement is at play where you are positioned. There is no 50/50; there is non stationary flow that is always determinant in the present with 100% certainty. The variable set on each fractal produces and provides an interlocking market view (and decision making from analysis) in the form of binary vectors should one desire to choose to make use of the information in this mechanical format. Because markets still have RTH's, it is the same as two conditions in the near term are operating. You can easily see this by looking at the two exclusive results of closes of RTH and closes (the ending price) of when the market is closed. One curve is growing and the other is falling. By looking at all trades, the MIT/BU group left out the delta of non RTH's which occurred as a consequence of holding overnight each night by necessity. what is the effect of bridging trading period to trading period? Anyone can see that this condition of markets precludes using random generators that can score very high on reality. Someone made the cursory comment that there is not too much gapping on most generator curves. I like the consideration of times when the market is changing orders by rule versus the times when this is not happening. It is also worth examining the marginal case of going from one to the other. I trade news by assigning values to news. In this way I have a standard trading strategy shift to take into account these three situations when they are present. I use a standard Greenspan as has been presented in the past. Price change through the application of market rules to change one kind of order to another is a cause of multisigma events on occasion. Being in the right side of the market is important here and the market operating point migration is important to track before during and after such events. This the case of "bad" news meaning unexpected often. Bad days are often the very best days in terms of market offerings. Market rules being applied to change orders from one kind to another are often present during BAD days that are very good days for someone who makes it a point to stay on the right side of the market. A Black Swan for you and Taleb is not a Black Swan for me in terms of trading strategies and the result of using my strategy.
Rahula Here is a "what if" for you. "What if", you for a moment, could take on the viewpoint that predicting is not a requirement to make money. To do this you would have to, for a moment, skip all the mathematics related to analyzing the markets from a probability point of view. Say you could for a moment. could you then just write out a list of how to do a routine to trade the market? As it stands right now, you are building a lot of stuff on some rather incomplete understandings of CW. I am not asking you to do something difficult like converting from an entry/ exit mode of existance to one of hold and reverse. you are too far gone to consider doing that. The deep negative feelings you keep getting about staying in a position need to be countered with critical thinking at some point.
human psych is not predictable in the micro which is all that is relevant in the second prior to decision. the above statement is irrelevant,misleading, and naive' it is often used but rarely questioned in the TA world by the true believers. everyone is trying to game the system, making the these trend ideas ancient legend. read "equity markets in action" for market facts not myth. surf
I'm not a frequent poster (quite obviously) or an experienced trader. I also could not read carefully all 25 pages of posts so hopefully I'm not repeating something mentioned earlier... with that out of the way, the random debate has always intrigued me and I wanted to run something by the people of this board. It's fairly easy to understand that one chart, on its own, is pretty much random. Yes, with Excel you can make a chart that's random. But randomness alone can't explain everything. There are too many people involved, to much psychology, to much information. The way I see it, the part that's not random is what moves on any given day and the relations between instruments. How could I tell a news trader that symbol XX is moving up 'randomly' when overnight earning news put it on the gappers list pre-open. I agree making money off the move a whole different story but the fact remains, the move is predicted with a certain accuracy that probably beat the 'random' odds. Same for sectors: the fact that an entire sector moves the same way for certain periods takes the randomness out does it not. In the Excel example done earlier, if you repeated the process 100 times, you would get 100 true random unrelated results. But I can look at the Dow, the NASDAQ, the S&P the eMini futures and tons of the other charts and see the same movements and make correlations between them. That doesn't feel random to me. Just my .02