Do you see patterns in Random Walks?

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

  1. MAESTRO

    MAESTRO

    Well, I am back with the brand new 4 inch scar right between my shoulder blades that is still oozing blood a bit. However, the pain is gone and I can type. I can tell you that lying on your stomach for a week is not fun at all! :mad:

    One thing I would like to mention. Interpreting the flocking behavior too literary is not going to work. The trick is to understand how to project the notions of the flocking behavior and spontaneous sync into the market's attributes. I suggest to look into one dimensional flocks at the beginning where the "birds" could only fly up or down. It might help the initial interpretation.

    Cheers,
    MAESTRO
     
    #651     Oct 17, 2011
  2. MAESTRO

    MAESTRO

    Yeah, I have heard many times about this research. All they do is Latent semantic analysis http://en.wikipedia.org/wiki/Latent_semantic_analysis
    on the internet data. It could be fruitful if the data has good representation of the phenomenon in question. In terms of the "formal financial education" I think it is irrelevant to this particular research. Many of very talented researchers in this field have never had any formal education in finance. I think it could be an advantage rather than a handicap.
     
    #652     Oct 17, 2011
  3. To me this is a load of crap MAESTRO. Someone can activate 100 robots in twitter posting all sorts of crap and the result is useless. There is no way to derive the validity of conclusions from premises in logic unless you know the validity of premises.
     
    #653     Oct 17, 2011
  4. MAESTRO

    MAESTRO

    I am not saying it is not :cool: I have tried this approach myself with no tangible results. I wouldn't say it is "a load of crap" but I would agree that it is difficult to see any reliable results in it. But, who knows ... It could be that with further development it might yield some positive results. I am still keeping my eye on this research.
     
    #654     Oct 17, 2011
  5. MAESTRO

    MAESTRO

    Also, a friend of mine who is in Germany and knows Tobias Preis personally has communicated with me recently. His opinion is very similar to yours. :D But I am not judging ... :cool:
     
    #655     Oct 17, 2011
  6. He's a young guy trying to market himself and his company, so maybe he does have some publications that serve that purpose. But, he hardly seems like a slouch-- Equalizer mentioned some his background. What do you guys think of his 'switching' papers? It reminds me Olsen's et al's market quakes research and Maestro kinetic energy tidbit. I will say his lit. review is lacking big time.
     
    #656     Oct 17, 2011

  7. The "realism" depends on the generating algorithm.

    If one simply generates random walk using GBM (as many people simply do), of course it's far from reality.

    These models have * constant * volatility, which you dont see in the mkt. Simulation must be less naive to make more sense.

    Tom

    PS
    check out the price simulator included here:
    http://www.datatime.eu/public/gbot/GBotScreenshots.htm
     
    #657     Oct 18, 2011
  8. MAESTRO

    MAESTRO


    I don’t think the question here is whether one could mimic the actual price behavior using any sort of a random generator or not. I think the more fundamental question here is whether the price behavior it self is a reflection of an objective random character of the market’s underlying forces or the price is a reflection of some sort of a deterministic non-linear algorithm that produces chaotic moves in the market.

    My personal opinion has always been that RANDOMNESS is an objective and necessary characteristic of nature. It is more persistent than we might think and governs many of the behavioral patterns that we (by being egocentric) mistakenly attribute to our conscious efforts. Accepting the fact that randomness exists on any level of organic and non-organic systems one could start fruitfully exploit its general laws (such as reversion to the mean) and investigate what happens when multiple random processes spontaneously synchronize. Randomness, actually, offers much more tools and insights for the research of a collective behavior phenomenon then any other models. Even fractals and their role in our day-to-day life is a part of this fundamental point of view.

    Cheers,
    MAESTRO
     
    #658     Oct 19, 2011
  9. SunTrader

    SunTrader

    Much of what has been said here I either don't understand fully or see much value in but this I do get and agree completely. :)
     
    #659     Oct 19, 2011
  10. In epistemology nobody cares about what we believe but how we justify our own beliefs. As such, I have not seen any convincing justification of such belief about randomness and it sounds to me more like religion or dogma than anything else.

    If as you say dear MAESTRO "RANDOMNESS is an objective and necessary characteristic of nature", why do things you lay on a table stay there unless disturbed by external forces? Why the arrow of time persists in only one direction and we do not see time reversals? Why things fall to the ground with an acceleration function that has been measured to be invariant of mass in one part in 15 trillion?

    Maybe at some level there is randomness but it is a mystery how this order we measure in the macroworld emerges from randomness. The obvious suspicion is then that this randomness is epiphenomenal and not a primary phenomenon, IMO. But again, I can't justify this belief.
     
    #660     Oct 19, 2011