Visual back testing of multi time frame indicators can happen only if you do visual back test and if your ea is set to work on - Technical Indicators

vito 2014.02.08 08:32 #621

I did already in runtime and happen what I described in the post( https://www.mql5.com/en/forum/general) I probably was not clear in my expostition ,I said " But, illogically (for me..), the visual arrow signal (steady mode) on 1Mtf graph (...sorry I meant>runtime) appears before the 5M tf bar closure and it's the same arrow that, at the same time, appears on a 5Mtf graph (>runtime) with identical indicator and with identical set of the prior one on the 1Mtf graph ". The graph (1m and 5m)are synchronous but on the first the signal appears in advance compared to the same signal of the 5m graph. Waiting for developments.. Thank again everybody.

mladen:
vitoingletto

You can not expect that a 5 minute indicator behaves the same way as 1 minute ones. 5 minute indicators have to display 5 same values over 1 minute charts (since 5 bars of 1 minute charts belong to same 5 minute bar)

Also, if you are using that test in a back test, in a back test indicators "know the future" and you can not use the current bar value as in runtime. Please test a multi time frame in a runtime and you will see exactly what is going on with multi time frame indicator on a current opened bar of the higher time frame indicator value

Request Execution - Opening Fractals - Bill Williams' Fractals - Bill Williams'

Mladen Rakic 2014.02.08 08:49 #622

vitoingletto:
I did already in runtime and happen what I described in the post( https://www.mql5.com/en/forum/general) I probably was not clear in my expostition ,I said " But, illogically (for me..), the visual arrow signal (steady mode) on 1Mtf graph (...sorry I meant>runtime) appears before the 5M tf bar closure and it's the same arrow that, at the same time, appears on a 5Mtf graph (>runtime) with identical indicator and with identical set of the prior one on the 1Mtf graph ".

The graph (1m and 5m)are synchronous but on the first the signal appears in advance compared to the same signal of the 5m graph. Waiting for developments.. Thank again everybody.

vitoingletto

If the indicator is set to use real time signals and it uses the signals from an opened bar it can not happen that you have a delay. What you are describing (that the signal on 5 minute indicator appears on an opened bar and does not change and that the ea uses that signal only after 5 minutes ) can happen only if you are doing visual back test and if your ea is set to work on a closed bar.

Please, avoid making conclusions based on visual back testing of a multi time frame indicators

Request Execution - Opening Signals of the Indicator Signals of the Indicator

vito 2014.02.08 09:07 #623

Thank for the quick answer. I'll see again on monday in runtime mode. thank again

mladen:
vitoingletto

If the indicator is set to use real time signals and it uses the signals from an opened bar it can not happen that you have a delay. What you are describing (that the signal on 5 minute indicator appears on an opened bar and does not change and that the ea uses that signal only after 5 minutes ) can happen only if you are doing visual back test and if your ea is set to work on a closed bar.

Please, avoid making conclusions based on visual back testing of a multi time frame indicators

Mladen Rakic 2014.02.08 10:01 #624

vitoingletto:
Thank for the quick answer. I'll see again on monday in runtime mode. thank again

I have a feeling that we are talking about different things all the time. That why I really do recommend that you test that in a run-time - that should clarify better some things than any explanation will

Structures and unions Data input Editing, compiling, and running

vito 2014.02.10 09:18 #625

Here we are, back again. Hope the picture is sufficiently clear of what I am trying to say. The same with identical set graph indicators give both arrows signal with 5 minutes delay each other. But the EA and the mail signal works (indifferently on M1&M5 GRAPH) according the M5 signal. That's a real pity because we lost, in this case, the first 5 minute earlier the signal when the movement is stronger and more robust. Please note that both M1&M5 indicators are in steadymode. There is any way to modify the indicator/EA to make them working on the first signal (the very first one om M1 graph )? A 2 cents suggestion. Would be already enough to make the EA works on the color change.... thank for your attention. Ad maiora!

mladen:
I have a feeling that we are talking about different things all the time. That why I really do recommend that you test that in a run-time - that should clarify better some things than any explanation will

Files:

vq_5mtf.jpg 405 kb

Determining Expert Advisor parameters Fractals - Bill Williams' Fractals - Bill Williams'

vito 2014.02.11 18:02 #626

update. Actually, on higher timeframe (M30-H1-H4), the indicator even in steady mode can repaint sometimes. But higher tf are usefull for understanting trend. Mladen was right. The lower tf seems to be more reliable..

vitoingletto:

Here we are, back again. Hope the picture is sufficiently clear of what I am trying to say. The same with identical set graph indicators give both arrows signal with 5 minutes delay each other. But the EA and the mail signal works (indifferently on M1&M5 GRAPH) according the M5 signal. That's a real pity because we lost, in this case, the first 5 minute earlier the signal when the movement is stronger and more robust. Please note that both M1&M5 indicators are in steadymode. There is any way to modify the indicator/EA to make them working on the first signal (the very first one om M1 graph )? A 2 cents suggestion. Would be already enough to make the EA works on the color change.... thank for your attention. Ad maiora!

Real and Generated Ticks Average True Range - Chaikin Oscillator - Oscillators