Announcement

Collapse
No announcement yet.

MACDx-over.efs

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Hello Jason,

    Thanks for the upateD link

    Here's an example (lower winDow) of how the single MACD stuDy with both histogram anD curves forces the histogram to be narrow (presumably to accomoDate the common zero line anD scale)

    However overlaying the stuDies (miDDle winDow) gets arounD this, but results in Discontinuities when Data is live

    I've cheateD here a little bit to clarify the Difference by ignoring the first extreme MACD histo bars of last FriDay's chart, but still the Difference is there

    RegarDs
    shaeffer
    (PS, please excuse the capital 'D's but I use "D" alone without Alt or Ctrl as a hotkey, which Disables small case 'D's - only shft D is available unless I Disable hotkeys
    Attached Files

    Comment


    • #17
      Hello again Jason

      In that efs, is it correct that I woulD only have to change
      12, 26, 9 in line 10 to my preference of 5,35,5, anD that var vMACD12_26 is only a label that woulD not have to be changeD throughout?

      That's slick, that the stuDy itself Does not have to be visible to get the alert (so it seems)?

      Thanks again
      shaeffer

      Comment


      • #18
        Hello shaeffer,

        In that efs, is it correct that I woulD only have to change
        12, 26, 9 in line 10 to my preference of 5,35,5, anD that var vMACD12_26 is only a label that woulD not have to be changeD throughout?
        Correct, but I would recommend opening the formula in the Formula Wizard to make the changes.

        What you are trying to do with the scaling issue isn't really a solution to your problem. The outputs of the histogram and the two lines are determined by input parameters to the study (5,35,5). By splitting this into two overlayed studies with the same parameters, you're not changing the outputs, only the scale. In your overlayed image, the two studies are not sharing the same axis, which is why to see the disconnectedness. They are plotting independent of each other, which allows the histogram to appear larger. The maximum and minimum values of each study determines it's own scale on the y-axis. If you go to Edit Studies and set each study to Scale Right and Display Right, you'll see that you're right back where you started.

        Without changing the MACD parameters for the histogram study, the only way to get "larger" histogram bars would be to multiply the histogram output by some multiple. This needs to be done through the EFS of course. In the return statement at the end of main add something like * 3 to the histogram return value.
        Jason K.
        Project Manager
        eSignal - an Interactive Data company

        EFS KnowledgeBase
        JavaScript for EFS Video Series
        EFS Beginner Tutorial Series
        EFS Glossary
        Custom EFS Development Policy

        New User Orientation

        Comment


        • #19
          Hi Jason,

          I went ahead earlier today and changed line 10 to my 5-35-5 parameters by opening the efs file and editting. I think it worked, but I also made some color changes and spent half the trading day chasing syntax error messages. (Loading a new efs into only one chart first and making one change at a time is my lesson there). On top of that, it seems the Triggered Alert List is giving me Up alerts when MACD crosses down - I'm trying to figure out why.

          I'll try using the Formula Wizard as you suggested (one more thing to learn!). I tried following the link in eSignal University on EFS Basics, but keep getting a lesson on ticker windows??

          Luckily I found the Guide to Developing eSignal Indicators in .doc format in the Forums (thanks Brad), because I ran out of paper after 20 pages, trying to print it from the EFS KnowledgeBase - I don't know why there was so much white space. So, I'm working my way thru that guide right now.

          How does the check box work when editting an efs..... does it look for the last case where the alert condition was met, or just run thru the first alert condition in the efs?
          ---------------------------------
          I didn't realize that having 'no scale' checked on my histogram was what allowed the larger histogram bars. Maybe an EFS with a *3 bar height factor will work, but as long as its linked to the MACD curves on a common y axis I'm picturing days where the curves might be in the upper 2/3 of the window, and the Histogram in the bottom 2/3, so neither would fill the window, a lose-lose result. And if *3 works for MXIM today, will it work for EBAY today, or MXIM tomorrow, or will I be having to change that to suit each day and each symbol?

          I'll think about your *3 suggestion, but really, as part of eSignal recently creating the feature to allow indicators to overlay each other, I think the ability for both indicators to fill the window should have been included. I use this feature, but maybe its not so commonly used to justify the fix. The same issue will come up with other overlayed indicators also (ie OBV and MoneyFlow).

          Anyway, back to the EFS Guide
          Thanks again for your reply
          Regards
          shaeffer

          Comment


          • #20
            MACD on price pane

            Originally posted by JayF
            At this point, there is no easy way to incorporate a MACD study into a price pane due to the way the Y-axis scaling works. We are working on enhancing eSignal, so that this can be done. Hopefully, this will be in the next version or two.
            Did you add this functionality to eSignal yet? I used the MACD this way in Tradestation.

            Thanks,

            Cindy
            Last edited by Cindy; 05-15-2005, 06:31 PM.

            Comment


            • #21
              MACD on Price Pane

              Ah, saw how to do this on another thread. Thanks!!

              Comment

              Working...
              X