Build Neural Network Indicator in MT4 using Neuroshell

Hi arryex,

Many thanks for your time... I started with NSDT and NS2 a few days ago and now is very clear for me how I can create an indicator in NS2 to NSDT !


Thanks again.

Clarod :clap:
 
Does anyone use entropy indicataor from noxa? can they post a screen shot how it looks like on daily euro chart? thx in advance
 
You can find the same entopy indicator in official mql4/5 website.Despite all that hype I could not make anything meaningful out of entropy indicator, hurst exponent, hölder exponent.
 
You can find the same entopy indicator in official mql4/5 website.Despite all that hype I could not make anything meaningful out of entropy indicator, hurst exponent, hölder exponent.


Tovim thanks .. is it the same enttropy on noxa because they mentioned entropy cannot be negative
 
I dont' know who is right who is wrong about normalization, foretrade got his formula from John Conover site and I find this entropy chart from Noxa very subjective.
 

Attachments

  • entropy.png
    entropy.png
    21.7 KB · Views: 424
Master Arry

I am tested recurrent networks for prediction fx moves. My whole data set has 1000 bars and i am training the network on 300 bars . would it be wise to leave inputs as it is i.e 0.5 for all weights ? The reason for this is i do not want to fit the data . All i want is recurrent network to learn the 300 bars pattern which i think is crutial for future moves
 
Folks was trying to extract eigen vectors from smaller time frames like 5 mins and looks it very difficult to get a eigen vector which is circular. any suggestions? this is using no xa software
 
There is a free MT5 neural network EA.
Can find it's name if you are interested.
 
Hi Supremegizmo,

You have a free choice to set the data range in NN as optimization, testing and production (out of sample). To avoid overfit, it is not necessary to limit the weight at a constant value. The most preferred parameters to verify the best NN model by identifying its performance in all above data ranges.
If you found that your model is only looks good during the optimization but not good during testing(paper trade) and out of sample, means your model is overfit.
I think the best way to check is by stopping during optimization and see the result in other data range. In NSDT we can only see the model performance after we stop the optimization, while in CH we can see the model performance in all data ranges in real time.
In your case, if you fix the weight in certain NN configuration, means you have only bias parameters in all NN layer that are used to approximate your output. hence, it is not wise to fix the weight value in your NN model.

QUOTE=supremegizmo;2199446]Master Arry

I am tested recurrent networks for prediction fx moves. My whole data set has 1000 bars and i am training the network on 300 bars . would it be wise to leave inputs as it is i.e 0.5 for all weights ? The reason for this is i do not want to fit the data . All i want is recurrent network to learn the 300 bars pattern which i think is crutial for future moves[/QUOTE]
 
Hi Supremegizmo,

You have a free choice to set the data range in NN as optimization, testing and production (out of sample). To avoid overfit, it is not necessary to limit the weight at a constant value. The most preferred parameters to verify the best NN model by identifying its performance in all above data ranges.
If you found that your model is only looks good during the optimization but not good during testing(paper trade) and out of sample, means your model is overfit.
I think the best way to check is by stopping during optimization and see the result in other data range. In NSDT we can only see the model performance after we stop the optimization, while in CH we can see the model performance in all data ranges in real time.
In your case, if you fix the weight in certain NN configuration, means you have only bias parameters in all NN layer that are used to approximate your output. hence, it is not wise to fix the weight value in your NN model.

QUOTE=supremegizmo;2199446]Master Arry

I am tested recurrent networks for prediction fx moves. My whole data set has 1000 bars and i am training the network on 300 bars . would it be wise to leave inputs as it is i.e 0.5 for all weights ? The reason for this is i do not want to fit the data . All i want is recurrent network to learn the 300 bars pattern which i think is crutial for future moves
[/QUOTE]

thx arry
 
Also wanted to know the dll files which can be used as indicaors on neuro shell which language are they codded c++ or c sharp? I am trying to code an indicator which is in mt4 to neuroshell
 
There are several samples available in ward.net, created using C, visual basic or power basic.

My self using C since not too much different with mq4.

Also wanted to know the dll files which can be used as indicaors on neuro shell which language are they codded c++ or c sharp? I am trying to code an indicator which is in mt4 to neuroshell
 
Top