Probabilistic Perceptron Neural Network

New comment
 

Some crazy idea just pop up in my head creating a Probabilistic Perceptron Neural Network

this is how it works.

every new bar, it will open a buy or a sell

the perceptron will adjust the weights and learn

give me your thoughts

How to use

Use the trainer to train,

inputs 0 - 200, step 1 for all

then sort by profit factor

 

Most suitable settings

doshur:
Some crazy idea just pop up in my head creating a Probabilistic Perceptron Neural Network

this is how it works.

every new bar, it will open a buy or a sell

the perceptron will adjust the weights and learn

give me your thoughts

How to use

Use the trainer to train,

inputs 0 - 200, step 1 for all

then sort by profit factor

Doshur,

Thanks for the perceptron,I have just downloaded it,so my comments are preliminary.

By looking at this code:

double ATR = iATR(NULL, 0, 13, 0);

double StopLoss = ATR * 3;

double TakeProfit = ATR;

I suggest that you modify it to

double ATR = iATR(NULL, 0, M,0);

double StopLoss = ATR * N;

double TakeProfit = ATR*P;

This way the perceptron will be able to find the most suitable ATR period and ATR multipliers for both stoploss and Take Profit.

I would use a range for M,N,P from 0.5 to 10 in steps of 0.5...as an alternative, you could make them external doubles and then optimize,but I believe that the fun of using the perceptron is for it to do the optimization on a continuous basis.If this is too wide a range and it slows the process ,you cold then make them external doubles,optimize to find where the best settings range for several different pairs and then reduce the "range of search" of the perceptron accordingly(if,for example,trough optimization of several pairs,you find optimal M(ATR period) ranges from 5 to 8..optimal N(Stoploss ATR multiplier) between 0.5 and 2.5..and optimalP(Atr multiplier for take profit) from 1.5 to 5.0..you can use those limits and program the perceptron accordingly).

Simba

 

actually im not try to find the ATR, TP and SL

rather i can input the weights to the indicator and aids in manual / ea trading as some filter

 

Did you took it from the Artificial Intelegence EA

he is using the same method.

 
myaron:
he is using the same method.

It doesn't really matter where I took it from because the implementation of perceptron is the same, weight X input ...

i only made some modification to it to be a Probabilistic one.

 

can you tell me when will the train EA produces the suitable result? also, what's the use of the perceptron_close indicator?

 
richest:
can you tell me when will the train EA produces the suitable result? also, what's the use of the perceptron_close indicator?

u can try on H1 or H4 for a few weeks

the indicator is for visual check on the chart (remember to input the optimize settings)

 

Have you or anyone tried membrain?:Membrain NeuralNet - MQL4 forum

 

Perceptron in action, see the forecast period. Seems pretty isn't it?

Files:
perceptron.jpg  195 kb
 

Be careful with what you're seeing there. The weighted-fitting approach is actually forecasting the past. What you need is the correlation between a clearly defined span of out-of-sample future compared to the last weight adjustment using a certain amount of latest historical data.

 
oR4z0r:
Be careful with what you're seeing there. The weighted-fitting approach is actually forecasting the past. What you need is the correlation between a clearly defined span of out-of-sample future compared to the last weight adjustment using a certain amount of latest historical data.

Understand that. This is just a test. If its working, I just add this as a filter of signal in my EA rather than depend on this to build my EA.

12
New comment
Reason: