Geez breakeven I wish you would have popped up yesterday. Are you referring to this exact same problem where the 15 min ES shows no volume when using the JHMVolume indicator but the 10 min and 30 min do? So you are saying that if I was showing the usual maximum of 15 days, when I cut it back to 12 days, the volume will reappear? That would be great if we are talking about the same thing. Thanks for your interest. lj
Thats it The problem seems to revolve a bit for me. Have had it happen on the 15min and 5min ES and sometimes on the 2min YM. The weird part is that it seems to be a certain day that causes the problem for me. Say the problem starts today, add 1 day to fix it...tomorrow I will sometimes have to add another day OR subtract 2 days. Sounds strange i know but seems to work for me Also, since the volume will show up when changing timeframes it may have to do with the number of bars on the chart (just a guess). I stopped investigating after finding the fix by loading more days. No guarantees, but it is worth a try. Hope it helps!
Right or wrong, I learned the IBGS as a shift, a reversal and yes, coded it that way. It is a very simple change: in the IBGS logic section, the 1st code line after the "if Low[0] < Low[1]" is to remove the GSL on the CurrentBar-1 (ie, previous bar). similarly on the High. If you don't want that then simply edit the Ninjascript, & comment out the two Remove stmts by placing a // at the beginning of the line. NT script editor will turn the line green when it is a comment. Then re-compile and re-apply the indicator to the chart (you do not have to remove & re-add, just click apply. BTW, the bar color is an NT problem. Barcolor in my code is set by the Volume indicator. Right after determining & assigning the volume bar color, if the chart is not candlestick style, then the price barcolor is set to the same color choice as the volume was. I have noticed, if you've just changed to bars (vs candlesticks) the volume indicator has to be applied again before the price bar colors are set. One of those many NT quirks. Sorry for the delay, I am not at home and have very limited access right now.
Thanks LT for your followup. I appreciate that it is the volume which is of consequence with respect to the 'real' bar color. My programming skills are still at thick as a brick level so if you wouldn't mind I'll show what I think you said, here, before actually trying it out. If I'm understanding what you said, I have 2 ways to fix this and one of them is to: comment out the two Remove stmts by placing a // at the beginning of the line. NT script editor will turn the line green when it is a comment. Then re-compile and re-apply the indicator to the chart (you do not have to remove & re-add, just click apply. Thanks again, lj
I think you are right with the assumption that this is a IBGS and IBGS is a SOC. but it appears to be in a place where we can not look for change, we are just in the start of a down traverse and bar 3 closes in the range of bar 2, so no SOC as the next bar confirms. The same holds true for the other 2 bars you are pointing to, in the opposite direction. P.S. I think this post is supposed to be discussed in 'Iterative refinement'.
An IBGS may or may not be a signal for change depending on the context and that is clearly demonstrated here. As well, you need to define what you mean by a signal for change in the sense of which fractal you are referring to, i.e., tape , fftraverse or traverse. All these points have been discussed on the IR thread and were not a part of the question which I was asking LostTrader. My question to LostTrader was not so much to do with the nature of an IBGS but rather why his coding appeared to 'discriminate' between varieties of IBGS bars depending on what the bar following the IBGS did. This as he has explained was a coding decision. This thread is particularly concerned with software and by implication the code which comprises a particular piece of software. lj
Hey breakeven, the freakin' thing fixed itself [maybe it has a built-in autorepair, ho, ho]. Anyhows I was talking with the NT crew and they have a couple of suggestions which I've placed in an attachment. lj
Hi LT, If you're still up to tolerating my programming ineptness, I've inserted a jpeg file which has my query about which 'remove ...' to comment out. TIA lj
Only the first one (highlighted in attached image) in each section that references the (previous) CurrentBar-1. Not the ones associated with the else stmt. Those are on the CurrentBar itself and you need to keep them to get any.
Thanks LT for your help. I will do the modification over this weekend and post the findings on this thread. Regards, lj (Laurent)