Hi Aroon,
I loaded up the indies on my MT4 yesterday and have gone back a lil bit into history.
This stuff is pretty amazing.
I dont know the performance in the long term but what I see is pretty good.
I will be testing this in the days to come.
Thanks once again for your efforts.
could you please tell us how to train ?
Arun,
I have noticed the neurotrend to repaint upto 4 bars back.
Is it because its taking its signals from 1 hour on the 15 min chart?
If so is it possible to remedy that as backtests in this case are not what they seem?
Please if any one can advise me on this ill be grateful.
Thanks.
Sorry to ask you this, but how come you find this ? its Saturday, and market isnt moving !
FYI :
from my knowledge this indicator can not load in to VISUAL MODE
do you mind to upload some photo as well ?
repaint like crazyyyyyyyyyyyyyyyyyy !
Sorry jayazan.
but I could not believe something “NEURAL NETWORKYYYY” can repaint ,.,
The repainting issue i noticed was on friday just before the markets closed…will post an image later…
Here you are. These are from today.
The first image shows where the neuro trend lines cross the 5 ema (aqua) to the upside, signaling a long entry at the end of the closed bar.
The second image clearly shows the repainting indy at “work”.
Yes i have noted and replied you in 165#
It was kinda hard and unbelievable for me to see something called neural network repaint…
Absolutely worthless like zigzag.
Thanks anyway.
I’ve seen your system only now. I’ve applied it today to the EUR/USD on a 15m chart. What happens if the four lines are completely undistinguishable ?
I know this thread is pretty old already and this is my first post here 
BUT when I searched on Forex NN I came here I found the reading pretty interesting but I can clearly see most of you approach this ‘approximation problem’ from the forex side and might (or might not) see Neural Networks as a ‘holy grail’ to forecasting. Actually it can be but that aside.
First of all, I am (only) 25 and have about 1 year experience in forex trading(like from 3 years ago when I had time to much). But now I do a Masters study in Intelligent Systems and with that AI like ANN are of daily order.
Did anyone ever wondered about the actual structure of the NN? I mean:
Does it contain hidden layers (like a must for approximation!!!) if yes how much? 2 hidden layers(its is considered 1 could approximate almost all functions) could be possible for approximation since financial problems are really complex.
If so what learning algorithms does this NN use? Standard back propagation? Mathatan learning rule? Resilent training? or maybe even: levenberg quadrant?(this would be for candle sticks rather than EMA).
What activation functions are used? naturally if you do not normalize the inputs it would be linear and output aswell but what was used for the hidden layer?
How are the inputs selected? Just by trial and error? or based on personal experience?(not bad though) or are ALL possible inputs tried and these came out as the best? (in matters of checking the NN weights and see if the NN ignores them as they provide no good information)
If there is some answer on those questions I would LOVE to know them!!!
Greets Olaf
Hello,
Is there an EA designed around this concept? Do you know how to use Avida together with MT4?
thanks for sharing.
does the system can be imply to all pairs?
tks
its kinda repaint itself
dont waste your time
arun bhai i hav 1 problem , i can see only one gold line that too 5ema nothing else please help very curious about the system
I see no indication that indicator values change after the bar is closed. I read the 5 digit values, and when I check the bar later, they are still the same. I think your eyes are fooling you.
Hi, I know this thread looks dead by now, but I’ll try my luck anyway. I’ve tried to build and use NN with the similar architecture (layers,inputs) as aroonraj suggested, but I chose antoher tool called FANN. Everything looked fine, until I tried to use FANN on a custom indicator to predict 1 bar future EMA. Well, it does predict something but the accuracy is rather bad. For instance if actual 15M EMA is 1.3925 on EUR/USD the NN predicts around 1.3975. 50 pips apart is useless on a 15M frame, so I wonder what is the problem? Is it FANN which performs worse than SNNS, is it the training process, or have I made some great fault somewhere. So if aroonraj or anyone else is still prepared to help a bit on this it would be great. Thanks in advance.
Hey A, I train or fine tune my trading brain every 3 months, my fiscal year starts in Sept, and I see the need to make the largest adjustment, from the end of June to the middle of September. Also the Western holidays. So maybe this might help in training your project, or maybe not hehehehehe. Anyway hope it works out for you.
The Ever Full Of Ideas VIPER
Hello Arun, Saw your old post on the NeuroTrend. Could you please suggest which development environment and libraries did you use to develop the NN system? I will appreciate your response.
So I’m in this pub with my mate - we’ve had a bit of a solid Sup up - like,
Pints of Milk of Amnesia all round …