Do you see patterns in Random Walks?

Discussion in 'Technical Analysis' started by atlTrader666, Aug 10, 2011.

  1. You confuse interpretations and experimental equivalence. All QM interpretations are experimentally equvalent. Knowing which one is true is equivalent to knowing the true nature of reality, which we do not. For now then, saying that the world is random because QM says so is a fallacy. That is all, no more, no less.
     
    #391     Sep 1, 2011
  2. MAESTRO

    MAESTRO

    Of course it is! The only question is which philosophical beliefs have broader horizons. Classical physics and relativity theory are also philosophical positions, however, certain things do not fit into classical mechanics anymore.

    here is the question. determinism assumes the ability to predict/anticipate certain outcomes of an experiment with a good degree of confidence. That is why we call it "determinism". However, can you conceivably imagine conditions where the outcome cannot be predicted with any degree of certainty? If yes, then there is inherent RANDOMNESS in that set of conditions, if NO, then all the outcomes of tomorrow's activities should be determinable today, but we know it's not the case.
     
    #392     Sep 1, 2011
  3. MAESTRO

    MAESTRO

    Thank you. let me think though this. I now see another degree of your point of view which is very interesting. Thank you for sharing. This is definitely a food for thought. Clearly, I need to do some more reading. Interesting thought, I have to admit.
     
    #393     Sep 1, 2011
  4. Samsara

    Samsara

    Incorrect. My point is <i>you</i> confuse interpretations and experimental equivalence. Note you used two words implying falsehood: "fallacy" and "misinterpretation". This is absolutely critical.

    If you had understood everything I'd said about Thomas Kuhn and Pragmatism, you'd realize my entire point is and remains that what is true is solely a product of our systems of verifying it as such, while cohering to everything else we hold as true.

    This is precisely why I feel it is fruitful to debate this exact issue.

    In order for you to avoid a pure logical contradiction you need to explain the process by which you deem interpretation X as a fallacy and misrepresentation.
     
    #394     Sep 1, 2011
  5. Interesting questions. I think I have answered them already based on my opinion. Determinism does not mean we must know with confidence. Determinism is a philosophical stand. To know, you have to know the model. To say that something was unexpected to you and therefore is random may be a naive stand. It just may be the case that you did not know how to predict it, the model.

    As far as tomorrow activities, they deal only with initial conditions and the model. Someone knows those initial conditions. When there were only pits for example, the locals knew all the initial conditions and could predict the market direction very accurately. Now, the exchanges and HFT know all iceberg orders and know where the market is going to go. You and I do not know because you and I do not know the model and the initial conditions. To you, randomness is an illusions. Someone always knows.
     
    #395     Sep 1, 2011
  6. I think I have to stop here because it appears you play with words.

    Bye
     
    #396     Sep 1, 2011
  7. Samsara

    Samsara

    Putting it as clearly as possible, before I basically have to agree with Mike805:

    The interpretation of QM that posits state reductions is a misrepresentation of the conclusions derived from the relevant experiments.

    True or false?

    Maestro's theory of randomness evident in market behavior relies upon the premises of the above interpretation.

    True or false?

    You already know by now I've agreed with you the entire time, save on the very first point.
     
    #397     Sep 1, 2011
  8. Intradaybill - I think this discussion about randomness is very poorly framed. Who actually believes in the market returns is ~N(0,svol)? Certainly not academics: the ones dealing with market efficiency usually deals with the fact that it's surprisingly difficult to system reject market returns is ~N(0,svol);

    In fact, a vast sub-sect of modern finance academia is in dealing with market structures that causes it to be NOT purely random.

    Finally, there are degrees of randomness. I don't think we should dwell on discussions about pure randomness because no one here nor in academia is actually dealing with that.
     
    #398     Sep 1, 2011
  9. I'm glad to see the discussion has evolved a bit and some posters have put aside petty ego conflicts. Maestro, glad to see you are well and sparking discussion again.

    Regarding Chaos, I've done some studies in this area and came to the conclusion it is not chaotic in the strict sense (deterministic).

    I've run numerous tests to test for chaos (and there are many), most fail. The one visual map that does show some type of non-randomness is something called a compass rose. It shows some underlying structural order which can show effects of price clustering but not determinism. Variance ratio tests from statistics show similar results.

    From a fractal point of view, I think everyone accepts that many financial series fail to be characterized by a hurst exponent of .5 (which implies random walk), and we also know that markets have an inherent positive drift over the long run (how long, who knows, all we have is history); neither implies determinism. Mandelbrot said markets are 'self affine,' which agrees with antisyzygy's (cool name btw), observations on different scales, but doesn't buy much in the way of deterministic prediction.

    All that being said, I think I read another poster's comment that a rat in a maze fared better than human subjects, as it converged to waiting in the box with 60% likelihood of getting fed. In many ways, I've argued buy and hold has an inherent edge built in, simply because of the positive drift and optimal net commission. Yet, it is still indeterministic and stochastic. Yet another way to think about edge existing-- even in the presence of randomness.

    Keep in mind, many of these comments relate to viewing properties from a time series perspective. Obviously, there exist front running and other cause-effect relationships in the market that may have highly predictable modeling properties, depending on the observer; but I'm operating under the assumption that the far majority of traders would never have sufficient access to these observables to make such a refined model (if you will). Therefore, much like classical physicists, we are stuck making models based upon aggregate ensemble properties of the information.

    One other side comment regarding chaos and entropy: I tend to think that markets organize in such a way as to optimize (maximize?) computational complexity, such that they continue to survive in the presence of invaders (opportunists, speculators, etc). In many ways, our bodies have evolved in the same manner (fractal, highly complex); it is when both systems have some moments of order (crash, tachycardia, seizure) is when they are most vulnerable and in some sense, most predictable. It's possible that the flocking behavior is another way of looking at this.
     
    #399     Sep 1, 2011
  10. MAESTRO

    MAESTRO

    Good to see you again! Great points! However, I think that it might be useful to agree on what RANDOMNESS is first before we go too far in analyzing its characteristics (or lack of). Don't you agree? I will try to come up with my own definition and will put it for discussion. I think it might clear a lot! Too often people argue about a subject having in mind totally different notions.
     
    #400     Sep 1, 2011