Something very simplistic

Discussion in 'Journals' started by Quah, Sep 11, 2002.

Thread Status:
Not open for further replies.
  1. dbphoenix

    dbphoenix

    Of course, it's only been three weeks . . .

    --Db
     
    #761     Oct 4, 2002
  2. "Only fools and dead people don't change their minds."

    Programming a method on TS hardly translates as a change in philosophy. Even executing the trades doesn't translate as a change in philosophy.

    Geez, can you tell I'm at home taking care of my kids today?

    :)
     
    #762     Oct 4, 2002
  3. dbphoenix

    dbphoenix

    Instead of waiting for Quah to do all this (assuming he's interested, which he may very well not be), much more progress would be made if those who are using this system were to run their own tests. For example, one might try various sto settings, either faster or slower or with different signal line lengths. Or one might test it with something other than sto. Or one might plot the same sto on both the 1m and 2m and enter only if they both agree as to direction. Or offset the entry by a bar. Or two. Or try every fifth bar. Or use the close of the bar previous to the entry bar rather than the open of the entry bar.

    Everybody, I assume, would like to avoid SARs and DWs. The question is how. And that means testing, boring though it may be.

    --Db
     
    #763     Oct 4, 2002
  4. I PM'd the same thought to Quah yesterday Db. I agree with you totally. The one I am looking at this weekend is taking the trade in the direction the %K is pointing, regardless of its position relative to the %D.
     
    #764     Oct 4, 2002
  5. dbphoenix

    dbphoenix

    I should mention that someone or ones was looking at using the direction of the previous bar for a direction signal, as well as using the relationship of the %K on the 1m and 2m as a signal, e.g., going long if the %K on the 2m had a higher value than the %K on the 1m, regardless of the value of the %D. I don't remember anything being reported about any conclusions, tho.

    --Db
     
    #765     Oct 4, 2002
  6. zorak

    zorak

    Quah, both trades I made today, I went the opposite
    way you did (9:33 and 9:43). They were both coinflips
    at the time...or so I thought. If the lines are nearly
    intersecting or otherwise very close, what do you use
    as a tie-breaker?

    Also, as someone else mentioned, it is incredible how
    frequently this situation seems to come up during these
    fib times.

    zorak
     
    #766     Oct 4, 2002
  7. Quah

    Quah

    The only way there could be a real tie is if they were both equal to the same value. I have my charting software set up to color my price bars based on the relationship of the stochastics lines - so I'm not even looking at the actual stochastics themselves. That resolves any issue in trying to figure it out visually.

    (edited to add: I'm using Ensign software from www.ensignsoftware.com - I don't work for them - doing something like this is very simple - there are probably 100 options on how to display indicators - one of them happens to be coloring the bars based on above/below - no code, just select the option.)
     
    #767     Oct 4, 2002
  8. dbphoenix

    dbphoenix

    As Quah has said, this is not an issue for him. If you don't want to use his software, tho, you can try using the same settings for both the 1m and 2m and not taking the trade if they don't agree as to direction. Whether you wait until the first bar where the direction is the same on both or wait until the next "scheduled" trade is up to you.

    --Db
     
    #768     Oct 4, 2002
  9. Wow, now that is cool.

    I just had a chance to look at the 10:59 trade for the ES. Looks like a good SAR. As was mentioned way back, the SAR operates fairly efficiently as a rule. Thanks to the Excel-lent records you are providing us, there appears to an opportunity there as a stand-alone.

    Ok good luck on the next one coming up.
     
    #769     Oct 4, 2002
  10. Okay I have the ES as a SAR B/E for the 1154 trade. Any discrepancies there?
     
    #770     Oct 4, 2002
Thread Status:
Not open for further replies.