Non-regular bars ?

Discussion in 'Technical Analysis' started by Swarm, Nov 30, 2010.

  1. First let me say that in the original explanation I mistyped, "1" should have been "2", "1" is not a prime number.

    Using primes seemed to be a reasonable place to begin since primes are used throughout information technology. A prime number (or a prime) is a natural number so it would be a natural (no pun intended) place to begin testing. I would be applying natural numbers to a non-variable environment, that is logical.
     
    #11     Nov 30, 2010
  2. What market do you usually trade?
     
    #12     Nov 30, 2010
  3. Vienna

    Vienna

    Proflogic: "Constant Volume Bars eliminate the arbitrary nature of price action because it makes news and other spike generating actions immaterial and irrelevant...."

    Jack Hershey: "To retain the most information, it is necessary to keep volume and price separate and of equal importance. The reason being that volume leads price. Signals come from volume and not only price."


    I have studied both. It is true that Constant Volume Bars make certain patterns easier to see...If you look at a CVB charts, it appears less "spiky" than a time based chart, so if your game is trading HL's etc., they seem to give a clearer picture. There are drawbacks though:

    The biggest one I have seen is that CVB's, by incorporating Volume in the bar or by eliminating time, remove the option of judging PACE, or the relative change of volume in an objective way:

    -For example: Strong moves occur on big volume. This is clearly visible if you have separate volume, but obscured if you do not.
    -Moves on declining Volume tend to fail. That is useful information,and better than relying on targets etc.
    -Moves that occur on huge volume but fail to reach important S/R (static or dynamic) can be an exit signal.
    -Breakouts without accompanying increasing volume fail.

    All these can be clearly seen and interpreted by comparing the bars in a separate volume pane but are much less clear when volume is integrated within the bar.
    (In order to get around these limitations, I even had someone write me an indicator a couple of years back that made a tick noise whenever a CVB bar completed...so I could actually HEAR the market 'speeding up'...since more volume bars printed in the same time-frame, that meant Pace was increasing....similar to Edmund Levefre listening to the sound of the ticker tape I guess. Did not use it though...)
    3. Jack has a point of course that picking a certain Contract amount per bar is no less arbitrary than picking 5 minutes for the length of a bar, and so is basing fractal nesting on this...always troubled me, just for reasons of logic... perhaps not a problem though, seems to work reasonably well.

    Volume bars per se do not "eliminate the arbitrary nature of price action", IMO. It seems intuitive to say they do because you seem to have eliminated TIME as a factor but I think this is a fallacy...

    Time based charts LOOK more chaotic, but that is really just a superficial appearance: I looked into the tool-sets that Jack's guys are using -for example, Sequences of Volume 'Gaussians' that occur at every breakout,'Pro-Rata volume' that tells you in the first minute of a bar where it is likely going to end up, channel lines that look like spaghetti to the untrained eye but are not once you can read them: if time-based charts were that "chaotic", price movement in them would hardly make it possible to use channels reliably...


    Having said all that, I still use Volume bar charts :).... I am used to them, figured out something that works with them, and I seem to have trained my Eye/brain to read them; however these days I have a broader view and am not anymore convinced that Volume bars are without question the most reliable way to present price movement on a chart.
     
    #13     Nov 30, 2010
  4. 4, 6, 8 and 9 are also natural numbers though.

     
    #14     Nov 30, 2010
  5. The bars are not the key to puzzle of reading price action consistently and successfully, they are the pieces of the puzzle. The key to the puzzle are oscillation tops and bottoms created on the chart. The pieces simply lead the way, bar by bar to the next preceding oscillation (key).

    I absolutely agree with your examples. Just goes to show you have learned to read a chart. Well done!

    What I do is not the be-all-end-all of charting. It is a consistent place to begin the learning process. It is a solid foundation where a trader can teach himself to become consistently profitable. You do not follow what I do in lock step but have taken the tools I gave you to make this wondrous environment your very own!!! The student has become the teacher.
     
    #15     Nov 30, 2010
  6. 4, 6, 8 and 9 are not least common multiples

    4 = 2x2
    6 = 2x3
    8 = 2x2x2
    9 = 3x3

    Primes are least common multiples

    2 = 2x1
    3 = 3x1
    5 = 5x1
    7 = 7x1
    11 = 11x1

    and so on and so on . . .
     
    #16     Nov 30, 2010
  7. This is the first time you mention LCM, is there a particular reason you chose natural numbers that are LCM's vs other natural numbers that are not? Thanks :)


     
    #17     Nov 30, 2010
  8. spd

    spd

    CL and TF are my favs. Also like 6E, NG, ES, and the grains.
     
    #18     Nov 30, 2010
  9. Post one of your five minute charts from today or tomorrow from Crude, Russell, Euro FX, Nat Gas, the S&P or any of the grains and I will post a volume bar comparison of the same area. Not til tomorrow though. It's bed time.
     
    #19     Nov 30, 2010
  10. I spent over a year testing chart increments and the best (visually appealing or clear) were those that were least common multiples . . . prime.
     
    #20     Nov 30, 2010