Complex Trend Detector - Indicator for MT4

Hello everyone !

A sub-unit of my company's main structure worked on a trend detector implementing the latest financial theories to see what was this all about.
It was inspired by the Physics Department's researches and mainly by the proprietary spectral filtration process they had elaborated.
A few months later, a sub-unit was created and they came up with this for the Algorithmic Unit.
This indicator finds the peaks and bottoms in entropy, filters them and smoothes them to finally analyze the cycle it is into.

The result (EURUSD H1) :



Here's my gift to you :



Make good use of it

Best regards


------- OLD FILES -------

Tanks you

Welcome, Dude !

Symphoenix,

Thanks for the indicator...

Could you go more into how to use it and any background info on the theory it is based on?

Dear Symphoenix,
Thank you for sharing your developed indicator. Couldn't really make any use of it for gold (XAUUSD), kind of shows the opposite and isn't available on the higher TF's. Would really appreciate some info on the logic it uses as quickly adjusting the parameters didn't correct for the opposite signal.

Kind REgards,
Tzamo

Thank you Symphoenix this indicator it repaint ?

Hello tzamo !

I think it's doing a pretty good job on XAUUSD H4 :



PS : First post updated with v1.1 as I've made a mistake translating python into mql4...

The beginning ot their journey started with this :





This helped them a lot too :

Thank you for sharing the indicator.
I want to use a chart illustration to show why I believe this indicator repaints. This is just from backtesting.

It repaints too much.

can you post without repainting? it seems to be good even with reapaiting but it takes much time to backtest it.. so if you post without repaint then it would be good. thanks

Wow.... if it was so easy, all of us would be millionaires in weeks. And I'm sure it's not so easy / possible to predict the market so precisely. All the pros and cons would come out on forward testing... in a matter of couple of days. Let's wait...!

And what does NASA has to do with this one? just being curious...!

Fascinating! ...and worthless. It repaints.

Houston, we have a problem:
Indeed it does. Save your money, folks.

At last we have proof that forex isn't rocket science.
______________________________

Here is an easy, objective way of checking the extent to which any indicator repaints:
(note: you can do this at any time, even while the market is closed)

1. Open a blank chart, and attach only the indicator you want to test to the chart.
Press F8 to get the Chart Options, and then ensure that 'Chart shift' and 'Chart autoscroll' (on the 'Common' tab) are both checked ON:
2. Save the chart as a template named tester.tpl:
3. Press Ctrl-R to open MT4’s strategy tester. Then enter the settings as shown in the screenshot below, i.e.

Expert Advisor: MACD Sample (we are simply using this factory supplied EA as a ‘dummy’ EA. We are not interested in the test results; we merely want to see the extent to which the desired indicator repaints)
Symbol: EURUSD (or anything you want, ideally it should match the pair/instrument of the chart that's currently open)
Period: ideally it should match the timeframe of the chart that's currently open
Spread: Current
Optimization: check this OFF
Model: Open prices only
Use date: check this ON, and select any suitable date range, e.g. from a few weeks ago, to today
Visual mode: check this ON. Move the slider to about speed 15 (the speed is shown in the little help bubble that pops up).
The Skip to date is irrelevant.

4. Click the ‘Start’ button to start the process. You can click the pause/play button (next to the speed slider) to pause/continue the chart video. Watch the progress and you can see if/how the indicator repaints. (Ignore any orders placed by the MACD Sample EA, obviously they are irrelevant).
______________________________

[EDIT] I find it curious how people ask for a repainting indicator to be re-posted without the repainting. Presumably they think that if a repainting indicator gives fantastically accurate signals -- picking tops and bottoms with amazing 'foresight' -- then a non-repainting version will be only marginally less valuable. Firstly, in some cases it's going to be an impossible task, if the repainting is part of the core calculation. And in other cases, it's the repainting that is the sole cause of the signals' seemingly fantastic accuracy. Remove the repainting and the indicator will lose its 'effectiveness' completely, and be reduced to the ranks of the also-rans (along with a zillion other indicators).

Perhaps it's a symptom of just how desperate people are to find the magic 'holy grail' indicator, one that will remove the uncertainty that's inherent in trading, and save them from the need to think through every situation for themselves. As a general rule, if an indicator seems to be able to pick tops and bottoms with amazing foresight, then it's almost certainly a repainter. No indicator can predict whether future buying volumes will exceed selling volumes at given levels (which ultimately determines if/how price will subsequently rise or fall).

Thank You Mr. Hanover for the detailed instructions. Learned something interesting today...

Can we have alert with this indicator Symphonix.

Interesting turn of events to say the least...

I just come here to share an indicator that proved to me to be useful and I'm being attacked on the fact that it repaints !

Hello, are there any Center Of Gravity users here ?

Repainting is NOT always a bad thing because the signals coming from this kind of indicators are ADAPTIVE...

You just have to know it to gain an edge. Like Hanover said, there is no holy grail, only pieces of it...

Best regards

Thanks for taking time to explain, Hanover!!! Worth reading.

Symphoenix...
Thanks for sharing the indicator. Not attacking you at all... sorry, if it sounded like that. And the points you made above is ABSOLUTELY spot on..! Thank you...

Ok, no 'attack' intended.
Perhaps I should say only smth like 'worthless for me'.
I mean, why not using smth simple, stable and effective as Pivot S/Rs+TLs?
Err, never mind, sorry ...I shouldn't say anything if your stuff doesn't suits me at all.
Good 'luck', all!