This question is for the math guru's

Discussion in 'Strategy Building' started by Sky123987, Nov 11, 2007.

  1. I'll confirm that MarkBrown has done it.

    Interesting thing about tick data derived signals are that, it serves as an OddBall-like signal. *In US index futures*

    Mark, how's your Cray computer?
     
    #11     Nov 14, 2007
  2. MarkBrown

    MarkBrown

    <p>This indicator was programmed over 10 years ago for Omega TradeStation.&nbsp; Here is a screen shot of an indicator which is placed on a 15 tick chart, that is each bar on the chart comprises of 15 transmitted ticks.&nbsp; The first screen shot <a href="http://www.oddballsystems.com/project/ob-028.gif">HERE</a> shows the 15 tick chart above and a special indicator which plots the range overtime.&nbsp;&nbsp;</p><p>The time interval can be varied and a screen shot of the time input can be seen <a href="http://www.oddballsystems.com/project/ob-029.gif">HERE</a>.&nbsp; It just so happens that I had the default setting set at 15 minutes as shown in the screen shot.&nbsp; I typically run this on a 5 tick bar chart with the time over range set to 15.&nbsp; I am displaying a larger frame tick chart so that the swings can better be seen.&nbsp;&nbsp;</p><p>Note the red lines shown on the indicator are a normal trading range and it is often interrupted with an above average trading range defined by the white line.&nbsp; The white line is a deviation based on volatility and then averaged to smooth it somewhat.&nbsp;&nbsp;</p><p>I appreciate the opportunity to display this to you because most of my work is often suppressed in many public forums.</p>
    <p><a href="mailto:markbrown@markbrown.com">Mark Brown</a></p>

    ps I am on my 3rd and fastest Cray ever. It is leased out during the day via remote terminal server to a engineering company and at night it's still all mine. Also I got it in Victory Red to match my C6 Z06 :D
     
    #12     Nov 14, 2007
  3. Given the assumptions in the original post, isn't the distribution similar to coin flips (heads +.01, tails -.01)?
     
    #13     Nov 14, 2007
  4. MarkBrown

    MarkBrown

    the only thing that can not be explained about the coin toss is the deviation and revision to the mean. even though the flips are thought to be independent if you keep count there are strings of runs which will revert. formal math teaching has failed us when it comes to these obvious anomalies.

    mb
     
    #14     Nov 14, 2007
  5. skepticaltrader

    skepticaltrader Guest

    Mark,

    You need to get rid of that slow Cray and get one of these.

    http://www.news.com/IBMs-Roadrunner...-marks/2100-1010_3-6218169.html?tag=nefd.lede

    :)
     
    #15     Nov 14, 2007
  6. MarkBrown

    MarkBrown

    blue is not my color although i look good in blue, i like red like the red in red storm cray.

    seriously i hand it to ibm they have really come along, but when you have history with something sometimes you just stick with it. besides seymour was a great eccentric guy. my aunt who has passed was a univac operator and i caught the bug that way.

    mb
     
    #16     Nov 14, 2007
  7. The outcome of a coin flip is not independent of outcomes from previous flips?
     
    #17     Nov 14, 2007
  8. MarkBrown

    MarkBrown

    i have proven it is not "at least to myself" and others have refuted it on the omega list years ago. however in practical application it works with such a high degree of regularity that it must exist.

    in fact by others own research - my theory works, how they can not rationalize. i have made alot of money using something that should not according to experts exist.

    now you can belive what you want. mb
     
    #18     Nov 15, 2007