Inverted head and shoulders on the SPX?

Discussion in 'Trading' started by fantastic4, Jan 28, 2009.

  1. This concept is fascinating! I took a stab at creating a CD database (cumulative delta for each tick level) based on the above example and then realized that I don't understand one key aspect of it.

    In the example above, we have -150k at 685.00 on March 3rd, -132k on March 5th and again -150k on March 6th. If the "determined" reference point is March 3rd, we're indeed net long +18k on March 5th and neutralized on March 6th. But what if we (hypothetically) also had -132k at 685.00 on March 2nd and that was taken as the reference point? In that case, we'd be neutralized on March 5th and net short -18k on March 6th. The key question seems to be how to determine the reference point. Anyone else sees that problem or am I not getting this right? Thanks again to AMT for the guidance so far.
     
    #831     Apr 1, 2009
  2. Nice study Gomi, thks.

    Where is the file located that stores the CD data and can it be viewed/opened w/ notepad?
     
    #832     Apr 6, 2009
  3. It's in My Documents, and its name is the instrument name.

    Please remember to use latest version which cures the 1 tick lag problem.
     
    #833     Apr 6, 2009
  4. rida07

    rida07

    Can someone please help me understand the following noob question:

    If market is quoted 814.25 - 814.50 and the following 3 trades happen:

    3@ 814.25/ 5@ 814.25 / 2@ 814.25

    Does this mean:

    1) There are 3 limit buy orders on the bid side for 3, 5 and 2 lots - and there was a 10 lot market sell order; or

    2) There is a 10 lot limit buy order, and there was a 3, 5 and 2 lot market sell orders?

    Is there a way to distinguish between the two?

    thanks
     
    #834     Apr 7, 2009
  5. There are AT LEAST 10 buy limits at 814.25

    3 separate sell orders came thru at the market, sell 3, sell 5 and sell 2.

    They all filled at 814.25 at the bids.
     
    #835     Apr 15, 2009
  6. Whisky

    Whisky

    Or alternatively 814.25 bid for 3 (order 1), 814.25 bid for 5 (order 2) and 814.25 bid for 2 (order 3) were the first in the GLOBEX2 queue and a seller sold 10 (or more) at 814.25,

    or 2 separate sellers sold 3, 5 ,2.

    or three separate sellers sold 3, 5, 2, or...

    Perhaps even more info (which clearing member had the buy orders and/or the sell orders and/or both maybe) can be gotten from a GLOBEX2 terminal. It used to be that way. Not sure now.

    JW
     
    #836     Apr 15, 2009
  7. Cme does not show counter party data any longer.
     
    #837     Apr 15, 2009
  8. Uncle

    Uncle

    thanks for the responses.. so in the market delta model - it's very difficult to get anything meaningful by filtering based on volume.. (as we are not able to tell whether the 3 volume was really on the buy or sell side), correct?
     
    #838     Apr 15, 2009
  9. Whisky

    Whisky

    Well, the sellers took the initiative to hit the bid. The sellers were the agressor. They might have been picking the wrong fight however, but at that moment in time the sellers CREATED the price at 814.25.

    In this method, as I understand it, you are trying to climb in the back (as a parasite?) of some bigger traders that are driving/ are going to agressively drive the price in the intended direction of your trade, because they have older inventory to liquidate, or they perceive the price to be an opportunity.

    I would like to create my own charts too...maybe one day. :p

    JW
     
    #839     Apr 15, 2009
  10. My math teacher in high school used to say that when he discusses new material and there's total silence, one of two things is happening: (1) everyone is in full understanding of the material or (2) nobody understands a thing...

    So, let me hammer on another issue I ran into with this example. Using uptick/downtick (with continuation) I am getting a CD differential of -45k between the LOD spike on march 3rd (it was actually 5 min into the next session) and the subsequent touch of that same price on march 5th. Assuming the +18k figure is *real* (and based on bid/ask method of CD calculation), is it possible that the uptick/downtick method would produce a contradictory reading of -45k?

    D.
     
    #840     Apr 16, 2009