{quote} Thank you for clearing this up for me trader2925. I believe have made the adjustments you mentioned and am now getting these results. Please let me know if these look correct or not. {image} EDIT: Uploaded stats with optimization = 1 and had to reupload with opt = .9 still doesn't really look right
{quote}. Maybe this pic will help you. edit: UR->? DJ should read 13 and not 3.So total is 148,not138. {image}
{quote} Thanks for trying to clear it up. So from what I understand all of the swings that have a T as the previous swing, we optimize T to use .7 of its lengths, and for R & J we optimize to use 1.1 of its length? Here is my result attempting to do this. I say attempting because if this is what we are doing, and my results are not matching there is a serious flaw in the logic of my code... {image} For some reason, I keep missing something in my code and can't seem to get the numbers to match
{quote} Maybe you've made the same mistake as I did first. Just to clarify... In fact every swing can have 3 different swing types at the same time: 1) Swing type with NO basic optimazation (RAW) 2) Swing type with basic optimazation (OBC) 3) Swing type with level 1 opitimazation (L1O) Probably 1 and 2 are correct, but to calculate L1O, you need to do the following: 1) Determine the OBC swing type from previous swing. 2) In function from this swing type you can select the right L1O parameter 3) Current swing type can then be determined with only...
Have finally caught up thanks to WannaLearn helping me find all the bugs that were messing up my code . I now believe I am getting the correct numbers. {image}
{quote} You'll have to wait for Rel's verification but that looks a lot better to me :o I think I've been having a similar problem to you, and my J swing percentages have been way too high for my liking (~20%). I'm wondering if there's a bug in my original code or something but I can't find it
Edit: I'm getting really confused as to how you calculate a current reject swing if you arbitrarily shrink the size of the previous swing.
Don't you need to keep the previous swing's end points to know if the current swing is a reject?
Do you always just take off 10% (assuming 90% optimization) from the top of the previous swing? (given that the 2 swings are previous D move and current U move)
First pic is by sys phus ,Second by me and third by LG.First 2 show very balanced DT-DR and UT-UR swings. {image} {image} {image}
{quote} I'll have a look, thanks! How do my raw swings look? Everything is a bit off so it's really hard to track down where exactly the problem is... Here are my swings with a .85 optimization..Which then throws the L1O way off balance. Final edit before i get out for a bit again: Once L1O is complete, do we track swing type to swing type via L1O swing to L1O swing or OBC swing to L1O swing? The original picture leads me to believe the latter but I'm not sure.. {image} {image}
{quote} Your reject swings seems to be on higher side(compared to reject swings in my raw swings data) .But i am not sure,it can be due to data duration you have used for your analysis.I have used 2016 data( jan-jun).You can check/compare your results with mine for this duration and locate the problem in your code. edit: {quote} It looks good.Is this result for .9 or .85 OBS classification?
{quote} Which pair did you use? Opps. That first picture is mis-labeled. Those results are for a .85 optimization rather than .90
{quote} Pair is EURJPY. {quote} That is interesting.Lets see if relativity confirms your result.
{quote} Once L1O is complete, do we track swing type to swing type via L1O swing to L1O swing or OBC swing to L1O swing? The original picture leads me to believe the latter but I'm not sure..
