Why do I see "Trends" in Randomly Generated Data?

Discussion in 'Data Sets and Feeds' started by Rahula, Feb 21, 2008.


  1. very nice, and from a very "high" location i understand :D


    [​IMG]
     
    #271     Feb 28, 2008
  2. Rahula

    Rahula



    Those charts show what brownian motion looks like then there are powerful underlying imbalances. In this case the powerful fundemental imbalances powering those moves was the rampant money supply growth in the past 5 years. And in the case of wheat, worldwide ethonal subsidies motivated farmers to plant corn instead of wheat resulting in a shortage of wheat.
     
    #272     Feb 29, 2008
  3. MAESTRO

    MAESTRO

    The latest technical analisys
     
    #273     Feb 29, 2008
  4. To the extent that you are using past market data to generate your trade criteria, then you are engaging in technical analysis. It may be of a different order than, say, the fringe that uses interplanetary lines of support, but it is technical analysis nonetheless. "Technical" because it has more to do with technique than it has to do with underlying economic considerations.
    :p

    As an aside, today and Wednesday were far better days for me than yesterday. How does that compare with your relative performance for those days?
     
    #274     Feb 29, 2008
  5. MAESTRO

    MAESTRO

     
    #275     Feb 29, 2008
  6. MAESTRO

    MAESTRO

    :D :cool: :D
     
    #276     Feb 29, 2008
  7. :D
     
    #277     Feb 29, 2008
  8. You're ornery.
     
    #278     Feb 29, 2008
  9. Don't forget the fact that the government of South Africa shut down the platinum mines, leading to a shortage of supply, leading to an imbalance of supply vs. demand.
    http://tinyurl.com/yrwmd8

    http://tinyurl.com/25s96y
     
    #279     Feb 29, 2008
  10. MarkBrown

    MarkBrown

    interesting, i have never seen this before - yet - within 10 seconds i was able to stop all the spinning by understanding why it was spinning and make adjustments to my focus and how i looked at it.

    i would say that looking at equity curves and mountains of data - i have also learned how to see patterns expertly.

    this "developed" ability has allowed me to focus on what works and what doesn't.

    there are 3 methods that absolutely work. they are:

    1.) reversion to the mean following abnormal volatility.
    2.) directional continuation supported by abnormal volatility.
    3.) range bound uncertainty or disinterest.

    why then hasn't anyone just killed the markets? well the math required to calculate what is normal volatility is tainted by the very subject that you seek which is abnormal volatility.

    and yet if you do master this math problem then the data which you apply it to "guess what" is also tainted by the introduction of poison which the name of is called _time.

    mb
     
    #280     Mar 1, 2008