Iterative Refinement

Discussion in 'Journals' started by Spydertrader, Jan 3, 2008.

  1. This is correct except for the "no wiggle room" part. If the market has yet to complete a sequence by the previous day's close then it must complete it the next day unless there is an overriding reason it cannot i.e. significant news release.

    In my experience this "continuation" is not a smooth transition from one day to the next. Perhaps due to "sync", but the market usually makes it interesting (read: not what one might expect) on its way to completing the sequence from the day before.
     
    #9531     Nov 24, 2008
  2. My un-confident view
     
    #9532     Nov 24, 2008
  3. Not likely, IMO PA can develop over the course of multiple bars just as easily as one bar, provided the volume difference is significant enough to place the volume bars in different zones delienated by pace lines. But, since I don't have a definition of Pace Acceleration, I can't really defend my argument here.

    In my defence, check the 11:20 bar on 10/31 - and the conversation in which Spydertrader so eloquently lead me (yet again) to the fact that everything one needs is in front of one's eyes.

    http://www.elitetrader.com/vb/showthread.php?s=&postid=2156707#post2156707

    http://www.elitetrader.com/vb/attachment.php?s=&postid=2155163
     
    #9533     Nov 24, 2008
  4. Actually, I've been looking at some charts, and this appears to be happening a lot after a lateral break out (directly or 2 bars after the breakout bar. Anyone else seeing this?

    --
    innersky
     
    #9534     Nov 24, 2008
  5. ehorn

    ehorn

    When you have a lateral (15:15-15:25), you wait for dominance to confirm. 15:30 was first bar out of lat (wait). 15:35 was DBV and internal (wait) plus it closes above the tape (tape from 14:55-15:00). 15:40 confirms dominance and allows one to seek a signal for change and 15:45 gives us Peak Volume and SOC.

    All times Eastern and [close-of] bar :)

    HTH
     
    #9535     Nov 24, 2008
  6. Dammit. I was debating whether or not to get into this again and thought WTF. I should have said in my proposal at least 3 bars.

    You are correct that Spyder very clearly says that the 11:20 bar shows pace acceleration and given that the pace levels are a catenary then it would seem that it is rigorous that the pace has in fact accelerated. I was going on the Gaussian slope definition which of course requires that at least 3 bars be involved.

    I checked the definition of the big "P" in TN's Spyder software and guess what? "P" = PEAK VOLUME. So now looking at the times one sees "P", what happens after the "P" is that volume falls off and rises again (this aside from any signal for change stuff). This can and frequently does happen when the pace levels are ALL extreme so there can be no zone transitions. There are also times when the two entities occur together.

    I do not have all these contextual matters resolved. However it can be said that by whatever means the pace changes arise, the consequences are similar.

    lj
     
    #9536     Nov 24, 2008
  7. Unfortunately, I can't really add anything to that other than "maybe."
     
    #9537     Nov 24, 2008
  8. The latter represents the same thought process as the former, with a slight variation. If something failed to complete the day before, then that something must reach completion, prior to, commencing the next phase of the process.

    - Spydertrader
     
    #9538     Nov 24, 2008
  9. You don't really think I would have made the solution as simple as downloading some software and looking at how I defined a variable do you? :) The answer doesn't reside in the code. The answer resides in the charts.

    - Spydertrader
     
    #9539     Nov 24, 2008
  10. I think we can do better than 'maybe' if Spyder is to be believed and I believe that to be the case.

    So pace acceleration is what he said it was in the earlier post referring to the "11:20" bar and peak volume is what he said it was based on his prior definition (acceleration of the Gaussian slope) with validation of this found in the definition of "P" in the TN software.

    He has never said that pace acceleration can be a signal for change but has said that under the proper circumstances, peak volume can be a signal for change - like today at 15:45 [the sequence was completed with 2 rising volumes, the second of which was a peak volume bar - sequence completion and signal for change on the same bar].

    FWIW, I see the after effects of peak volume and pace acceleration to be very similar - dominant volume falls and rises again. There may be more to this, as I have said, but for now I believe it is useful information. While not wanting to enter a delusional state, it does seem worthwhile to use the 'givens' to search for the 'unknowns' and not dump the givens because we have not eliminated the unknowns.

    lj
     
    #9540     Nov 24, 2008